PDF, Anglais, 175 pages, 3,6 Mo - Femise

In collaboration with:
Jawad Kerdoudi: IMRI, Morocco
Yazid Boumghar: CREAD, Algeria
F E M I S E R E S E A R C H
P R O G R A M M E
2008-2009
Convergence of Banking Sector Regulations and its
Impact on Bank Performances and Growth:
the case for Algeria, Egypt, Morocco, and Tunisia
Ce rapport a été réalisé avec le soutien financier
de l’Union Européenne au travers du Femise. Le
contenu du rapport relève de la seule responsabilité
des auteurs et ne peut en aucun cas être considéré
comme reflétant l’opinion de l’Union Européenne.
Research n°FEM33-04
Directed By
Rym Ayadi, CEPS, Belgium
July 2011
This document has been produced with the financial assistance
of the European Union within the context of the FEMISE
program. The contents of this document are the sole responsibility
of the authors and can under no circumstances be
regarded as reflecting the position of the European Union.

CONVERGENCE DES RÉGLEMENTATIONS
BANCAIRES DANS LE SUD DE LA MÉDITERRANÉE
IMPACT SUR LA PERFORMANCE DU SECTEUR
BANCAIRE ET SUR LA CROISSANCE
RYM AYADI
EMRAH ARBAK
SAMI BEN NACEUR
BARBARA CASU
AVEC LES CONTRIBUTIONS DE
MOHAMMED YAZID BOUMGHAR
JAWAD KERDOUDI
MOEZ LABIDI

The Centre for European Policy Studies (CEPS) is an independent policy
research institute in Brussels. Its mission is to produce sound policy research
leading to constructive solutions to the challenges facing Europe. The views
expressed in this book are entirely those of the authors and should not be
attributed to CEPS or any other institution with which they are associated or to
the European Union.
This report was directed by Dr. Rym Ayadi, Senior Research Fellow and Head
of the Financial Institutions and Prudential Policy research unit at CEPS. Other
contributors include Emrah Arbak, Researcher in the Financial Institutions and
Prudential Policy research unit at CEPS, Barbara Casu Lukac, Reader in
Banking at the Cass Business School in London and Sami Ben Naceur,
Associate Professor at the University of Tunis. Country expertise was provided
by Mohammed Yazid Boumghar, Researcher at the Centre de recherche en
économie appliquée pour le développement (CREAD) in Algeria, Jawad
Kerdoudi, President of Institut Marocain des Relations Internationales (IMRI)
in Morocco and Moez Labidi, Director of Research, Applied Economics and
Simulation, Faculty of Economics and Management in Mahdia, University of
Monastir, Tunisia.
The report was produced with financial assistance from the European Union
within the FEMISE programme 2008-09 (reference FEM33-04). Established in
June 2005, the Forum Euroméditerranéen des Instituts de Sciences
Économiques (FEMISE) is a Euro-Mediterranean network consisting of more
than 80 economic research institutes, representing the 37 partners of the
Barcelona Process.
ISBN 978-94-6138-086-9
© Copyright 2011, Centre for European Policy Studies and the authors.
All rights reserved. No part of this publication may be reproduced, stored in a
retrieval system or transmitted in any form or by any means – electronic, mechanical,
photocopying, recording or otherwise – without the prior permission of the Centre for
European Policy Studies.
Centre for European Policy Studies
Place du Congrès 1, B-1000 Brussels
Tel: (32.2) 229.39.11 Fax: (32.2) 219.41.51
E-mail: info@ceps.eu
Internet: www.ceps.eu

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
EXECUTIVE SUMMARY
La récente crise financière a de nouveau placé au centre du débat politique les standards et normes bancaires
internationales dans les pays développés. Le débat est loin d’être récent, et ne concerne pas exclusivement
les pays développés, puisqu’un système financier solide où les règlements sont correctement appliqués peut
aider les économies en développement à mieux canaliser leurs ressources financières vers l’investissement.
Pour les économies ouvertes, un tel système financier peut également agir comme rempart, facteur
aujourd’hui important face à la volatilité des marchés.
Avec ceci en toile de fond, cette étude analyse l’impact des réglementations bancaires sur l’efficacité du
système bancaire et la croissance économique dans 4 pays du Sud de la Méditerranée : Algérie, Egypte,
Maroc et Tunisie tout en explorant le niveau de convergence des pratiques réglementaires et de l’efficacité
bancaire par rapport aux standards Euro-Méditerranéens 1 .
Premièrement, l’étude compare les secteurs bancaires de l’Algérie, de l’Egypte, du Maroc, de la Tunisie et
leurs réglementations respectives par rapport aux standards internationaux. Pour ce faire, des mesures de
l’adéquation des pratiques de supervision et de régulation sont utilisées. Deuxièmement, l’efficacité bancaire
et le niveau de convergence sont étudiés en utilisant l’Analyse d’Enveloppement de Données (AED).
L’analyse est complétée par le recours aux méthodes de Meta Frontière, Beta et Sigma convergence.
Troisièmement, en utilisant les méthodes de mesure des pratiques de réglementation et de supervision,
l’impact de l’environnement légal sur l’efficacité des banques est étudié. L’étude prend en compte non
seulement le détail des différentes réglementations mais aussi les caractéristiques légales et institutionnelles
des pays du Sud de la Méditerranée. Quatrièmement, l’étude analyse comment l’adhésion à ces normes et
standards pourrait influencer le potentiel de croissance de chaque pays.
Le Chapitre 1 est une analyse descriptive des secteurs bancaires de ces quatre pays. Le Chapitre 2 développe
ensuite des mesures d’adéquation réglementaire dans plusieurs domaines et compare les pays du Sud de la
Méditerranée avec l’UE-MED. Le Chapitre 3 résume l’analyse de l’efficacité et de la convergence entre le
Sud de la Méditerranée et l’UE-MED. Le Chapitre 4 propose une analyse empirique des déterminants de
l’efficacité bancaire qui se concentre particulièrement sur les mesures d’adéquation réglementaire
préalablement développées. Le Chapitre 5 développe une analyse similaire pour la croissance économique.
Le dernier Chapitre conclue et présente les principales recommandations politiques émanant des résultats de
l’étude.
Cette recherche a été dirigée par Dr. Rym Ayadi, « Senior Research Fellow » et Chef de l’Unité de
Recherche sur les Institutions Financières et Politique Prudentielle au CEPS. Les autres contributeurs sont
Emrah Arbak, Chercheur dans l’Unité de Recherche sur les Institutions Financières et Politique Prudentielle
au CEPS, Barbara Casu Lukac, Professeur de Banque à la Cass Business School de Londres et Sami Ben
Naceur, Professeur Associé à l’Université de Tunis. L’expertise pays est fournie par Mohammed Yazid
Boumghar, Chercheur au Centre de Recherche en Economie Appliquée pour le Développement (CREAD),
Algérie ; Jawad Kerdoudi, Président de l’Institut Marocain des Relations Internationales (IMRI), Maroc ; et
Moez Labidi, Directeur de Recherche, Economie Appliquée et Simulations, Faculté d’Economie et de
Gestion a Mahdia, Université de Monastir, Tunisie.
Ce rapport a été produit avec le support financier de l’Union Européenne via le programme FEMISE 2008-
2009 (réf. FEM33-04). Etabli en 2005, le Forum Euro-Méditerranéen des Instituts de Sciences Economiques
(FEMISE) est un réseau Euro-Méditerranéen de plus de 80 centres de recherche représentant les 37
partenaires du Processus de Barcelone.
1 Les pays Euro-méditerranéens ici considérés sont Espagne, Chypre, Grèce, Italie, Malte et Portugal (ci après dénommés « UE-
MED »).
2

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
Chapitre 1- Aperçu des secteurs bancaires d’Afrique du Nord
Ce chapitre fournit une analyse descriptive des systèmes financiers dans les quatre pays Sud-Méditerranéens
considérés. Les rôles du secteur public et du cadre légal sont évalués qualitativement afin de souligner les
principales déficiences du secteur bancaire de chaque pays, ces déficiences allant de la structure des marchés
bancaires aux détails des bilans agrégés des banques, en passant par les indicateurs de sûreté financière.
L’analyse révèle plusieurs points communs entre les secteurs bancaires des pays du Sud de la Méditerranée.
Au cours des dernières années, les autorités bancaires des pays étudiés se sont engagées dans une série de
réformes visant à moderniser leurs secteurs bancaires. Celles-ci incluent la restructuration et la privatisation
des banques publiques, la mise en place de législations prudentielles et de systèmes de gestion des risques
ainsi que le renforcement des prérogatives de supervision. Le Maroc et l’Egypte ont amélioré la disponibilité
et le partage d’informations relatives au crédit. Ces réformes se sont traduites par la croissance durable du
crédit au secteur privé.
L’analyse montre qu’une des explications potentielles du sous développement financier réside dans le haut
degré d’intervention de l’Etat. Cet interventionnisme est aussi bien direct qu’indirect : il se manifeste par la
présence de banques détenues par capitaux publics, ainsi que par l’exposition à la dette publique des
portefeuilles des banques. Au cours des dernières années pour lesquelles les données sont disponibles (2008-
09) la part de marché des banques publiques oscille entre un quart du total des actifs bancaires au Maroc; 67
et 90% en Egypte et en Algérie respectivement. La structure de propriété des banques ainsi que des
conditions sous-jacentes comme les rendements élevés sur la dette publique en Egypte excluent
vraisemblablement le crédit aux entreprises privées. En effet, la dette publique et les prêts (y compris aux
entreprises publiques inclus) représentent à peu près un tiers des du total du bilan des banques Algériennes et
Egyptiennes, dépassant ainsi la part des crédits au secteur privé.
Mis à part l’exclusion du secteur privé du marché du crédit et les contraintes sur le développement financier,
le rôle dominant de l’Etat dans le secteur bancaire semble avoir un impact négatif important sur la qualité
même des crédits. En effet, les ratios de créances douteuses sur les créances totales dans le Sud de la
Méditerranée sont parmi les plus élevés du monde. Grâce à un interventionnisme moindre le Maroc reste une
exception avec le ratio de créances douteuses le plus bas parmi les pays étudiés. De plus, les quatre pays
étudiés ont mis en place des politiques visant à améliorer la qualité des créances comme de meilleures
dispositions dans la privatisation des systèmes d’information de crédit, des programmes de rachat des crédits,
ainsi que d’autres plans d’assainissement des bilans. Malgré cela, la qualité des crédits dans les pays du Sud
de la Méditerranée est la plus mauvaise parmi les pays de la région
La persistance de créances douteuses et de systèmes financiers sous développés conduit à se poser la
question de l’adéquation des réformes réglementaires des systèmes bancaires des quatre pays étudiés.
Comme il a été noté ci-dessus, la prépondérance des banques publiques pourrait bien être la racine du
problème. Toutefois, les défaillances des régimes légaux et de supervision pourraient aussi avoir leur
importance.
Chapitre 2 : Convergence des réglementations des systèmes bancaires
Le chapitre précédent a montré que tous les quatre pays ont mis en place des réformes de leurs secteurs
financiers lors des dernières années. Dans ce chapitre, grâce aux informations publiquement disponibles ainsi
qu’à des enquêtes sur les réglementations bancaires conduites depuis 2000 dans un grand échantillon de
pays, plusieurs indices sont développés pour évaluer et suivre l’évolution ainsi que l’adéquation des
réglementations bancaires. Afin de permettre des comparaisons dans l’ensemble des pays Méditerranéens,
les indices sont calculés pour un total de 11 pays Méditerranéens : 5 pays Sud Méditerranéens (Algérie,
Egypte, Israël, Maroc et Tunisie) et 6 pays UE-MED (Chypre, Espagne, Grèce, Italie, Malte et Portugal).
3

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
Le but de ce chapitre et de fournir des mesures quantitatives des évolutions réglementaires qui pourraient
servir d’indicateurs dans les exercices empiriques à suivre. Grâce à l’utilisation de différentes sources de
données, les sept domaines réglementaires suivants sont couverts afin d’en évaluer l’adéquation : la
définition de l’activité bancaire, les obligations liées à l’octroi d’une licence, les exigences en capital,
l’indépendance et le pouvoir du superviseur bancaire, la présence de filets de sécurité, la diffusion et la
disponibilité d’informations sur le crédit. Bien que ces domaines donnent une vue d’ensemble des
réglementations, plusieurs autres axes comme les systèmes de paiement et de compensation, les schémas de
garantie, et l’inclusion financière pour ne citer que ceux la, ont été exclus pour cause de manque de données
comparables entre pays.
Dans ce chapitre, la qualité de la réglementation et de supervision des pays Sud Méditerranéens ainsi que
leur convergence vers les standards UE-MED sont revues. L’évaluation a été conduite sur sept dimensions y
compris l’étendue de l’activité bancaire, les barrières à l’entrée, la sévérité des exigences en capital, le
pouvoir et l’indépendance de l’organe de supervision, les incitations fournies par les régimes d’assurance des
dépôts, la surveillance privée, le droit des créditeurs et l’accès à l’information.
L’évaluation collective de la convergence des structures de réglementation et de supervision des les pays du
Sud de la Méditerranée vers les standards de l’UE-MED présente un bilan mitigé. Malgré quelques
améliorations, des faiblesses persistent dans le cadre de l’assurance des dépôts, les barrières à l’entrée et
l’information sur le crédit. Par ailleurs, des disparités se sont manifestées récemment, particulièrement pour
ce qui est de la rigueur des normes prudentielles, du potentiel d’intervention politique et de la surveillance
par des acteurs privés.
L’indice d’assurance des dépôts ne s’est pas amélioré car ni les autorités Egyptiennes ni les Tunisiennes
n’ont mis en place un régime formel d’assurance des dépôts. Comme il a été souligné précédemment, les
garanties implicites peuvent favoriser la prise de risques grâce à une garantie couverte du gouvernement pour
les prêteurs. Par ailleurs, même en Algérie et au Maroc aucun effort n’a été fait pour aligner les incitations
des banques par la mise en place de primes gagées sur le risque ou de systèmes de co-assurance, ce qui
permettrait pourtant d’internaliser certains des coûts dus à une prise de risque excessive et encourus par les
régimes d’assurance des dépôts.
Les pays du Sud de la Méditerranée ont mis en place un certain nombre de réformes destinées à améliorer la
disponibilité et l’utilisation des informations sur le crédit par les institutions financières. L’Egypte et plus
récemment le Maroc ont mis en place des registres de crédit ou « credit bureaus » respectivement en 2006 et
2009, mais les différences entre le Nord et le Sud de la Méditerranée ne se sont pas résorbées. L’Algérie et la
Tunisie continuent de se baser sur les registres publics, les deux pays restreignent l’accès aux historiques de
crédit des emprunteurs, et aucun ne collecte ni ne distribue de données détaillées émanant de sources non
bancaires comme les commerces de détail ou les services publics. Bien que la littérature ne donne que peu
d’indications, les registres de crédit privés bénéficient d’un meilleur accès à de nouvelles technologies et
savoirs faires afin de s’assurer du fonctionnement correct des mécanismes de partage de l’information. Les
pays de la région devraient ainsi continuer à surveiller le développement de systèmes d’information à la
pointe, afin d’utiliser le stock de données déjà construit par les autorités publiques 2 .
Un autre domaine qui demeure une faiblesse majeure des systèmes de réglementation de la région est celui
des barrières à l’entrée. Bien que les exigences pour l’octroi de licences se ressemblent de part et d’autre de
la Méditerranée, d’autres indicateurs font état de d’importantes barrières à l’entrée. La présence du
gouvernent dans les structures de propriété du système bancaire, très répandue dans la région, donne des
2 Le Maroc pourrait servir d’exemple intéressant. Le pays a fusionné les fonctions et capacités de collecte de données de Bank Al-
Magrhib opérant sous le registre public avec celles du bureau de crédit privé récemment établi Experian-Maroc. Pour une analyse
comparative des systèmes d’informations marocain et égyptien voir Madeddu (2010, pp.21-23)
4

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
avantages injustifiés aux banques concernées et restreint les incitations à entrer sur le marché. Au Maroc, la
part des banques détenues par capitaux publics dans le total des activités bancaires est en déclin dans le total
des activités bancaires, tandis qu’en Algérie en Egypte et dans une certaine mesure en Tunisie la présence du
gouvernement perdure. Bien que la présence du gouvernement puisse donner lieu à des avantages indirects,
les autorités doivent en revanche s’assurer que les rôles soient bien définis et que la présence des banques
publiques ne représente pas un obstacle pour le développement du système financier 3 . Le taux de refus
auxquels ont été sujettes les banques étrangères sont très élevés, ce qui renforce l’idée de barrières à l’entrée
et d’avantages concurrentiels dont bénéficient les banques domestiques.
En sus de ces trois faiblesses majeures, en 2007 l’enquête a révélé 3 nouvelles sources préoccupantes de
défaillances. Les exigences en capital qui en 2003 étaient alignées sur les standards Européens se sont
détériorées selon la dernière édition de BRSS. Il existe quelques exceptions comme l’Algérie et dans une
moindre mesure la Tunisie. Toutefois, en Egypte et au Maroc, les exigences en capital et les règles
comptables sont devenues plus flexibles et moins sensibles au risque. De mauvaises pratiques comptables ont
également contribué à davantage de disparité dans les indices de surveillance privée.
Enfin, les quatre pays se caractérisent par un potentiel élevé d’ingérence politique dans le système bancaire
nuisant ainsi aux autorités de supervision et renforçant le contrôle direct du gouvernement –une source de
préoccupation supplémentaire pour la compétitivité et l’efficacité du secteur bancaire. Comme en témoignent
les soulèvements de 2011 en Egypte et en Tunisie, les gouvernements de la région ont essayé de garder, peut
être pendant trop longtemps, un contrôle étroit sur leurs systèmes politiques et économiques, or ce sont
précisément ces formes d’ingérence qui pourraient entrer en conflit avec les objectifs des autorités
financières et de la concurrence.
Chapitre 3 : Analyse de l’efficacité et de la convergence
Cette partie de l’étude tente d’apporter un peu de lumière sur ces sujets en examinant l’effet des réformes
financières sur l’efficacité du secteur bancaire dans 11 pays de la région Méditerranée (MED-11) pour la
période 1995-2008: Chypre (CY), Algérie (DZ), Egypte (EG), Espagne (ES), Grèce (GR), Israël (IL), Italie
(IT), Malte (M), Maroc (MA), Portugal (PT) et Tunisie (TN). La deuxième partie de l’analyse vise à
contribuer au débat actuel sur la dynamisation de l’intégration dans la région Méditerranée. Suite aux travaux
de Casu & Girardone (2010), nous utilisons les concepts de β-convergence et σ-convergence ainsi qu’une
analyse dynamique des données de panel pour évaluer la vitesse d’intégration des marchés financiers dans la
région.
Nos résultats indiquent une amélioration de l’efficacité bancaire dans la région, plus particulièrement dans la
dernière partie de la période étudiée. L’efficacité bancaire moyenne augmente, poussée par les
développements technologiques et la convergence des banques aux meilleures pratiques. Les banques
espagnoles dominent la région aussi bien pour ce qui est de l’efficacité totale que pour les ratios de métatechnologie.
Néanmoins, lors de la période étudiée, le ratio moyen de méta-technologie pour la région est lui
aussi en augmentation, indiquant ainsi une capacité des banques de la région à s’approprier les meilleures
technologies disponibles sur le marché.
Ces résultats sont appuyés par les estimations de la β-convergence et de la σ-convergence. La β-convergence
est toujours négative et statistiquement significative indiquant ainsi l’existence d’une convergence des scores
d’efficacité dans la région MED-11. De plus, les résultats de la σ-convergence suggèrent une augmentation
3 Rocha et al. (2010) soulignent le rôle essentiel joué par les banques publiques dans l’octroi de crédits aux PME. Les auteurs notent
que les banques privées sont incapables de combler ce vide principalement à cause de la mauvaise qualité de l’infrastructure
financière ainsi que de la disponibilité et de la fiabilité des informations sur les emprunteurs potentiels.
5

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
de la vitesse de convergence dans la mesure où le coefficient σ est toujours négatif et statistiquement
significatif.
Chapitre 4 : Impact des réglementations bancaires sur l’efficacité
Ce chapitre pose une question très spécifique: l’efficacité-coût des banques Méditerranéennes est elle plus
importante dans les pays caractérisés par de meilleures conditions réglementaires et de supervision ? Les
résultats font écho aux dernières découvertes relayées par la littérature. Certains aspects de la réglementation
comme les obligations de divulgation, la disponibilité d’informations sur le crédit et les barrières à l’entrée
sont très importantes. L’existence d’un régime formel d’assurance des dépôts peut aussi renforcer l’efficacité
en attirant l’attention sur le besoin d’améliorer la confiance des déposants. Les autres résultats sont moins
clairs et appellent à une étude plus approfondie. Par exemple, bien que les restrictions sur les activités portent
atteinte à l’efficacité, il est possible qu’elles conduisent simultanément à des risques plus importants.
Les résultats de ce chapitre montrent clairement que les banques opérant dans des pays à meilleure structure
de réglementation sont les plus efficaces. Particulièrement, l’existence d’un régime formel d’assurance des
dépôts, les obligations de divulgation qui facilitent la surveillance par les acteurs privés ainsi que la
disponibilité d’informations sur le crédit améliorent l’efficacité des banques. Dans une moindre mesure, les
résultats pointent vers un impact positif de la rigueur des exigences en capital sur l’efficacité des banques.
Par contre, selon nos résultats, il est possible d’autoriser les banques à étendre leurs activités en démantelant
les barrières à l’entrée du secteur. L’indépendance de l’organe de supervision semble ne pas avoir d’impact
sur l’efficacité coût des banques, un résultat vraisemblablement dû à un effet de compensation trouvant son
origine dans les risques élevés émanant de la centralisation du pouvoir politique.
En somme, nos résultats supportent le troisième pilier de Bâle II avec un plus faible support pour les
exigences en capital. Le déploiement rapide de registres du crédit privés, probablement calqués sur le modèle
marocain est également un facteur important de l’efficacité bancaire. Toutefois, aucun de ces facteurs ne
devait être traité isolément. La qualité institutionnelle, mesurée ici par une l’agrégation d’un ensemble de
facteurs politiques et de gouvernance est une caractéristique d’importance capitale. Enfin, la stabilité
macroéconomique contribue également de manière importante à l’efficacité des banques.
Il doit aussi être souligné que les résultats de ce chapitre ont également évalué l’importance des pratiques de
régulation et de supervision dans l’amélioration de l’efficacité bancaire. D’autres sujets devraient également
être considérés pour faire une évaluation plus large de la pertinence et l’adéquation de certaines règles et
standards.
Chapitre 5 : Impacts des réglementations bancaires sur la croissance
Ce chapitre se tourne vers une investigation plus profonde des bénéfices économiques émanant des pratiques
de réglementation et de supervision. La principale question est de savoir si les réglementations et pratiques
bancaires ont un impact sur la croissance. Plusieurs canaux par lesquels la relation peut opérer sont ici
considérés, y compris l’impact des réglementations sur l’efficacité-coûts du système bancaire, l’octroi de
crédit au secteur privé et l’activité sur les marchés de capitaux. L’analyse empirique prend également en
compte d’autres canaux intermédiaires qui ne sont pas considérés.
Selon nos spécifications, les réglementations financières ont un impact direct relativement limité sur la
croissance économique. Parmi les sept domaines réglementaires considérés, seul le contrôle par le
gouvernement – proxy des barrières à l’entrée et des conditions de marché – semble avoir un effet négatif et
régulier sur la croissance économique. De plus, les preuves de l’impact du développement financier sur le
développement économique sont limitées. Bien que l’efficacité et la rotation sur le marché des actions
6

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
semblent avoir un effet positif sur le taux de croissance du revenu par tête, le crédit au secteur privé apparait
comme ayant seulement un impact limité (et possiblement négatif) sur la croissance.
Les résultats montrent que les réglementations exercent un impact sur la croissance principalement à travers
des variables financières. Sur la base des résultats revus dans cette section, il apparaît que les facteurs de
réglementation examinés ici n’exercent au mieux qu’un impact indirect sur la croissance par le biais du
développement financier. De plus, l’absence de corruption a un impact clair sur la croissance ce qui souligne
son importance comme condition préalable à la croissance.
À ce stade plusieurs faiblesses techniques doivent être soulignées. Premièrement, la petite taille de
l’échantillon ne permettait pas une analyse en panel. Deuxièmement, les similarités entre les pays étudiés ont
pu se traduire par des biais d’échantillonnage, ce qui implique que les résultats doivent être interprétés avec
précaution et doivent faire l’objet d’une réévaluation avant d’être transposés à d’autres régions.
Troisièmement, la similarité entre les pays étudiés rend difficile la recherche d’indicateurs forts dans la
mesure où la variation inter-pays est limitée. Ceci est en effet l’une des principales causes de la faiblesse
apparente des instruments choisis pour la part du crédit au secteur privé rapportée au PIB. Enfin, on a assumé
que les variables de réglementation ne changeaient pas sur la longue période. Bien que cette hypothèse ne
conduise vraisemblablement pas à d’importants biais, elle est une source additionnelle d’homogénéité.
Conclusions
Cette étude met en lumière les changements dans les cadres réglementaires de quatre pays Méditerranéens :
Algérie, Egypte, Maroc et Tunisie. Lors des deux dernières décennies, les quatre pays ont conduit
d’importantes réformes du secteur financier de différente ampleur en termes de profondeur, engagement et
succès. Dans le cas du Maroc par exemple, la suppression des subventions et des contrôles des taux
d’intérêt ; le renforcement des rôles et mandats des superviseurs bancaires ; l’amélioration des pratiques de
gestion du risque et leur alignement sur les pratiques les plus récentes ; la mise en place réussie d’un régime
d’assurance des dépôts et la création d’un système d’information sur le crédit ont permis de construire le
système financier le plus développé parmi les pays ici étudiés et peuvent très bien servir d’exemple de
bonnes pratiques pour d’autres systèmes financiers en développement.
Les autres pays Sud Méditerranéens examinés ici ont moins bien réussi à mettre en place des réformes clés.
Comme le montrent les ratios élevés de créances douteuses, les banques des trois autres pays ont des actifs
de moins bonne qualité. Les politiques mises en place pour répondre à la mauvaise qualité des actifs se sont
traduites soit par de maigres améliorations, soit par un assèchement du crédit, voire par les deux. Les efforts
de privatisation n’ont été que partiellement réussis. Parfois, ces efforts ne sont pas traduits par une
amélioration des conditions de marché et de développement financier. En Algérie, les banques publiques
continuent de dominer le secteur bancaire et comptent pour 90% des actifs totaux du secteur bancaire. En
Egypte, bien que les efforts de privatisation aient été partiellement couronnés de succès, les prêts au secteur
public et la dette représentent une fraction importante des portefeuilles bancaires, grevant ainsi les
possibilités de développement financier ainsi que les opportunités de croissance. En Tunisie, la plupart des
grandes banques du pays sont détenues par capitaux publics.
Les comparaisons entre les pays UE-MED présentent des déficiences particulières. Malgré quelques
améliorations récentes, les barrières à l’entrée sont encore monnaie courante dans les pays Sud
Méditerranéens. Elles trouvent leur origine dans les taux élevés de refus d’octroi de licences aux banques
étrangères, eux-mêmes intimement liés à la part dominante de l’Etat dans le secteur bancaire. Les exigences
en capital sont moins rigoureuses dans les pays Sud Méditerranéens considérés, de plus en plus à cause des
disparités dans la sensibilité au risque des exigences en capital minimum. Les régimes d’assurance des
dépôts existants – c.à.d. ceux de l’Algérie et du Maroc – créent des incitations adverses et peuvent
augmenter les risques d’aléa moral. En Egypte et en Tunisie, les garanties implicites du gouvernement
7

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
pourraient également aggraver les problèmes d’aléa moral. Bien que la surveillance privée et les exigences
de transparence semblent alignées sur les standards UE-MED, les pratiques comptables sont de plus en plus
mauvaises dans le Sud de la Méditerranée. Enfin, malgré quelques améliorations récentes, la disponibilité
d’informations sur le crédit est plutôt mauvaise dans la région.
En ce qui concerne l’analyse de l’efficacité-coûts, les résultats indiquent une amélioration globale des
niveaux d’efficacité aussi bien dans les pays UE-MED que dans le Sud de la Méditerranée depuis 2005
(exception faite de l’Egypte). Pour le Sud de la Méditerranée, si cette amélioration est particulièrement
flagrante pour les banques Algériennes et Marocaines, elle résulte de facteurs différents. L’efficacité
moyenne globale des banques du Sud de la Méditerranée s’améliore, encore une fois poussée par les
améliorations dans les bonnes pratiques. Les banques UE-MED et en particulier les banques espagnoles
dominent la région avec des scores d’efficacité de l’ordre de 80.4% contre une moyenne régionale de
63.5%. Les banques espagnoles possèdent également les ratios de méta-technologies les plus élevés de la
région avec une augmentation constante de ces derniers. Ceci indique que les banques espagnoles ont
constamment amélioré leur technologie bancaire devenant un exemple de bonnes pratiques. Toutefois, durant
la période analysée le ratio moyen de méta-technologie est en augmentation ce qui indique que les banques
de tous les pays considérés sont capables de s’approprier la meilleure technologie disponible.
Ces résultats sont confirmés par l’estimation de la β-convergence. Le coefficient β est toujours négatif est
statistiquement significatif, indiquant ainsi qu’une convergence dans les scores d’efficacité des pays de la
région a bien eu lieu. Par ailleurs, les résultats den la σ-convergence suggèrent une augmentation de la
vitesse de convergence car le coefficient σ est toujours négatif et statistiquement significatif. Ceci montre
que si l’écart technologique est encore élevé, il tend à se réduire plus rapidement.
En examinant l’impact des pratiques de réglementation et de supervision sur l’efficacité coûts des banques,
les résultats montrent clairement qu’une bonne structure de réglementation contribue de manière importante
à l’efficacité du système bancaire. A cet égard, le cas du Maroc est instructif. En particulier, les régimes
d’assurance des dépôts, d’adéquates obligations de divulgation de l’information ainsi que la disponibilité
d’informations sur le crédit semblent améliorer l’efficacité des systèmes bancaires. Une définition plus large
du marché bancaire résultant de barrières à l’entrée moins importantes ainsi que de restrictions moindres
dans l’activité des banques améliore aussi l’efficacité des banques, mais à un degré moindre toutefois. Le
déploiement rapide des bureaux de crédit, vraisemblablement basés sur les meilleures pratiques régionales
améliore également l’efficacité comme le montre le nouveau système Marocain.
Enfin, l’impact positif sur la croissance de l’adéquation réglementaire semble opérer principalement à travers
son impact sur le développement financier. L’étude montre également que la présence du gouvernement dans
le système bancaire nuit à la croissance, et ce, même en dehors du cadre du développement financier et
d’autres variables financières. De plus, des obligations de divulgation plus exigeantes ainsi que des
exigences en capital plus rigoureuses et des limitations moindres de l’activité bancaire ont également des
impacts positifs sur la croissance en améliorant le développement financier.
Il est aussi important de noter que les pratiques de régulation et les facteurs d’adéquation ne doivent pas être
traités séparément. La qualité institutionnelle mesurée ici par un ensemble de facteurs liés à la gouvernance
et à la politique est un facteur important dans toutes les régressions. Le contrôle de la corruption et la
présence d’institutions démocratiques sont aussi des facteurs importants qui doivent être appréhendés
parallèlement aux conditions de réglementation.
En somme, l’étude met en lumière certaines des déficiences les plus importantes des réglementations
bancaires du Sud de la Méditerranée. Il apparaît que quelques uns des standards les plus nouveaux comme
les exigences en capital de Bale II ont été conçues pour les pays développés et ne semblent guère appropriées
à d’autres pays en développement à cause de nombreuses déficiences dans le partage de l’information, dans
8

Convergence des Réglementations Bancaires dans le Sud de la Méditerranée
les exigences de divulgation et plus globalement dans le cadre institutionnel. Un des objectifs clés des
prochaines réformes devrait être de trouver des moyens pour réduire l’emprise du gouvernement sur le
système bancaire tout en s’assurant que le cadre réglementaire et le développement institutionnel qui en
émanent répondent correctement aux imperfections du marché.
9

CONVERGENCE OF BANK REGULATIONS ON
INTERNATIONAL NORMS
IN THE SOUTHERN MEDITERRANEAN
IMPACT ON BANK PERFORMANCE AND GROWTH
RYM AYADI
EMRAH ARBAK
SAMI BEN NACEUR
BARBARA CASU
WITH CONTRIBUTIONS FROM
MOHAMMED YAZID BOUMGHAR
JAWAD KERDOUDI
MOEZ LABIDI
1

EXECUTIVE SUMMARY
I
international standards and norms on banking regulations have, once again, leaped to the
forefront of the policy discussion in developed nations due to the recent crisis in the world’s
financial markets. This discussion is far from new, nor does it apply exclusively to the
world’s most advanced economies. A sound and well-enforced regulatory regime can help
developing nations to channel financial resources more efficiently into investments. For open
economies, it can also act as a buffer, an important stability factor in today’s shaky market
situation.
Against this backdrop, this study examines the impact of banking sector regulations on
bank efficiency and economic growth in four Southern Mediterranean countries (referred to
collectively as “South-MED”) – Algeria, Egypt, Morocco and Tunisia – while exploring the level
of convergence of regulatory practices and efficiency to EU Mediterranean 1 standards.
In particular, the study first compares the banking sector and its regulations in Algeria,
Egypt, Morocco and Tunisia, to international standards using measures on the adequacy of
regulatory and supervisory practices. Second, banking efficiency and the level of convergence
to best practices are examined using Data Envelopment Analysis (DEA) and complemented by
Meta Frontier Analysis and the β-convergence and σ-convergence methodologies. Third, the
impact of the regulatory environment on the efficiency of banks is investigated using the
developed measures of regulatory and supervisory practices. In addition to the regulatory
details, the performance analysis also considers the legal and institutional characteristics of the
South-MED countries. Fourth, the study explores how compliance with these standards and
norms may influence the growth potential of each country.
Chapter 1 provides a descriptive analysis of the banking sectors of the South-MED
countries covered in the study. Chapter 2 then develops measures of regulatory adequacy in a
number of areas and provides comparisons with the EU-MED. Chapter 3 summarises the
analysis of efficiency and convergence between the South-MED and EU-MED. Chapter 4
analyses empirically the determinants of the efficiency scores, paying special attention to the
regulatory adequacy measures developed. Chapter 5 provides a similar analysis for economic
growth. The final chapter concludes and puts forth the main policy recommendations.
The research was directed by Dr. Rym Ayadi, Senior Research Fellow and Head of the
Financial Institutions and Prudential Policy research unit at CEPS. Other contributors include Emrah
Arbak, Researcher in the Financial Institutions and Prudential Policy research unit at CEPS, Barbara
Casu Lukac, Reader in Banking at the Cass Business School in London and Sami Ben Naceur,
Associate Professor at the University of Tunis. Country expertise was provided by Mohammed
Yazid Boumghar, Researcher at the Centre de recherche en économie appliquée pour le
développement (CREAD) in Algeria, Jawad Kerdoudi, President of Institut Marocain des Relations
Internationales (IMRI) in Morocco and Moez Labidi, Director of Research, Applied Economics and
Simulation, Faculty of Economics and Management in Mahdia, University of Monastir, Tunisia.
The report was produced with financial assistance from the European Union within the FEMISE
programme 2008-09 (reference FEM33-04). Established in June 2005, the Forum Euroméditerranéen
1 The European Mediterranean countries considered in the study are Cyprus, Spain, Greece, Italy, Malta
and Portugal (referred to as “EU-MED”).
2

des Instituts de Sciences Économiques (FEMISE) is a Euro-Mediterranean network consisting of
more than 80 economic research institutes, representing the 37 partners of the Barcelona Process.
Chapter 1- Overview of the national banking sectors in North Africa
This chapter provides a descriptive analysis of the financial systems of the four South-
MED countries. The role of the public sector and the changing regulatory and legal framework
are assessed qualitatively to highlight the main shortcomings of the banking system. Apart from
legal sources and international assessments, the discussion relies on quantitative measures of
the banking sector in each country, ranging from structure of banking, details on aggregate
balance sheets and indicators of financial soundness.
The foregoing analysis reveals several common features of the banking sectors of the
Southern Mediterranean countries. In recent years, the authorities of the four surveyed
countries have engaged in a variety of reforms to modernise their banking systems. These
include restructuring and privatisation of public banks, implementation of prudential
regulation and risk management frameworks and enhancing supervisory responsibilities.
Morocco and Egypt have improved the availability and sharing of credit information. These
reforms have led to a persistent growth of credit to the private sector.
The analysis shows that one potential explanation of financial under-development is the
heavy presence of the state, either directly in the form of publicly-owned banks or indirectly in
the form of public debt in banks’ portfolios. For the latest years for which data are available
(2008-09), the market shares of public banks range from a low of one-quarter of total banking
assets in Morocco and Tunisia to highs of 67% over 90% Egypt and Algeria, respectively. These
ownership structures and the underlying conditions, such as the high returns that government
debt earns in Egypt, are likely to crowd out the credit to private enterprises. Indeed, public debt
and loans, including loans to public enterprises, account for nearly one-third of the total balance
sheets of the Algerian and Egyptian banks, surpassing the share of private credit.
Aside from crowding out private credit and constraining financial development, the
state’s dominant role in the banking sector appears to have a serious negative impact on credit
quality. Indeed, the ratios of non-performing loans to gross loans for the Southern
Mediterranean countries are among the highest globally. Owing to the relatively limited role of
the state, Morocco is once again an exception, with the lowest NPL ratios among the four
countries. Moreover, the four countries have implemented policies to improve the quality of
loans, including privatisation improvements in credit information systems, loan repurchase
programmes and other plans to clean balance sheets.
The persistence of the non-performing assets and underdeveloped financial systems
remain leads to questions on the adequacy of the recent regulatory reforms in the banking
sector in the four countries covered in this study. As noted above, the prevalence of the
publicly-owned banks may be at the root of the problem. However, shortcomings in various
legal, regulatory and supervisory frameworks may also matter.
Chapter 2: Convergence of banking sectors regulations
The previous chapter has shown that all the four countries have faced substantial reforms
in their financial sectors in recent years. In this chapter, a number of indices are
developed in order to assess and track the evolution of the adequacy of banking
regulations using publicly available and comparable surveys on banking regulations for a large
sample of countries since the early 2000s. To allow comparability across the Mediterranean, the
section develops the measures for a total of 11 Mediterranean countries, including five South-
3

MED countries (Algeria, Egypt, Israel, Morocco and Tunisia); and six EU-MED countries
(Cyprus, Spain, Greece, Italy, Malta and Portugal).
The aim of this chapter is to develop quantitative measures of regulatory development
that could serve as an indicator in the empirical exercises that follow. Seven distinct regulatory
areas are identified for assessing different dimensions of regulatory adequacy. These cover
definition of banking, licensing requirements, capital requirements, independence and power of
supervisor, presence of safety nets, disclosure and availability of credit information using
distinct data sources. Although these provide a broad view of the extent of regulation, several
potential areas (i.e. payment and settlement systems, credit guarantee schemes, financial
inclusion, etc.) have been excluded due to the unavailability of comparable information sources
for the sampled countries.
This chapter reviewed the quality and the level of convergence of the regulatory and
supervisory structures of the South-MED and EU-MED. The assessment included seven
dimensions, including scope of banking, entry obstacles, the stringency of capital requirements,
the power and independence of the supervisory authority, incentives provided by the deposit
insurance scheme, private monitoring and creditors’ rights and access to information.
Despite some improvements, key weaknesses remain in deposit insurance, entry obstacles
and credit information. Moreover, some recent issues recent disparities have also become
apparent, especially in the stringency of capital requirements, potential for political interference
and private monitoring.
The deposit insurance index has failed to improve since neither the Egyptian nor Tunisian
authorities have put in place an explicit insurance scheme. As discussed previously, implicit
schemes may enhance risk-taking through a blanket government guarantee for the leading
institutions. Moreover, even in Algeria and Morocco, no effort has been made to align the
banks’ incentives by implementing risk-based premiums or co-insurance schemes, which would
help internalise some of the costs to the deposit guarantee schemes due to excessive risk-taking.
The South-MED countries have implemented a number of reforms to improve the
availability and use of credit information by financial institutions. Egypt and, more recently,
Morocco have established private credit bureaus in 2006 and 2009, respectively. Nevertheless,
the South-North gap has not been narrowed. Algeria and Tunisia continue to rely only on
public registries, restrict the borrowers’ right to inspect their credit histories, fail to collect and
distribute detailed data, including from non-bank sources, such as retail stores or utility
companies. Although the literature provides little guidance, private credit bureaus have an
improved access to new technologies and know-how to ensure that information-sharing
mechanisms work effectively. The countries in the region should continue to monitor
developments and spearhead innovative systems to use the stock of information and
infrastructure already set-up by the public systems. 2
Another major issue, the presence of entry obstacles, continues to be a key weakness of
the regulatory structures of the region. Although the licensing requirements exhibit similarities
on both sides of the Mediterranean, other indicators point at substantial barriers to entry.
Government ownership, which is widespread in the region, gives undue advantages to
2 Morocco may serve as interesting example, by effectively combining the data collection roles and
capacities of the Bank Al-Maghrib, which operates the public registry, and the newly established private
credit bureau, Experian-Morocco. For a comparative analysis of the Moroccan and Egyptian credit
information systems, see Madeddu (2010, pp. 21-23).
4

incumbent banks and restricts entry incentives. In Morocco, the government-owned banks
represent a declining proportion of total bank activities; in Algeria, Egypt and (to some extent)
in Tunisia, the government-ownership persists. Although government ownership may have
some beneficial side benefits, the authorities have to ensure that such roles are well-defined and
should not be an obstacle to the development of the financial system. 3 The rates of foreign
denials are also very high, further supporting the idea of substantial entry barriers and
competitive advantages enjoyed by domestic incumbent banks.
In addition to these three key weaknesses, the 2007 survey points at three new concerns.
The stringencies of capital requirements, which were in line with the EU standards in 2003, have
deteriorated according to the latest BRSS survey. There are some exceptions, like Algeria and, to
a lesser extent, Tunisia. However, in Egypt and Morocco, the capital requirements and
accounting standards have become more flexible and less risk-sensitive. Poor accounting
practices have also contributed to an increasing disparity in private monitoring indices.
Lastly, political interference has become a significant possibility, potentially undermining
supervisory authority and reinforcing the governments’ direct control – an additional concern
on the competitiveness and efficiency of the banking sector. As the eruption of public discontent
in Tunisia and Egypt in early 2011 clearly attests, the region’s governments have attempted to
maintain (perhaps for far too long) a tight grip on their countries’ political and economic
systems. It is exactly such forms of interference that may conflict with the objectives of the
financial and competition authorities.
Chapter 3: Analysis of efficiency and convergence
This part of the study attempts to shed light on these issues by examining the effect of
financial reform on the efficiency of the banking sector in 11 countries in the
Mediterranean region: Cyprus (CY), Algeria (DZ), Egypt (EG), Spain (ES), Greece (GR),
Israel (IL), Italy (IT), Malta (MT), Morocco (MA), Portugal (PT) and Tunisia (TN) over the
period 1995-2008. The second part of the analysis aims to contribute to the current debate on
fostering integration in the Mediterranean region. Following Casu & Girardone (2010), we use
the concepts of β-convergence and σ-convergence and employ a dynamic panel data analysis to
assess the speed at which financial markets are integrating.
Our results indicate an improvement in bank efficiency across the region, particularly in
the latter part of the sample period. The overall mean efficiency in the region is increasing,
driven by technological improvements by the best practice banks. Spanish banks dominate the
region both in terms of overall efficiency and of meta-technology ratios. Nonetheless, during the
sample period, the average meta-technology ratio for the region is also increasing, thus
indicating an ability of banks in all countries to appropriate the best available technology.
These results are supported by the estimation of β-convergence σ-convergence. The β
coefficient is always negative and statistically significant, thus indicating that convergence in
efficiency scores has occurred across countries in the MED-11 area. Furthermore, results for the
σ-convergence suggest an increase in the speed of convergence as the σ coefficient is always
negative and statistically significant. This indicates that, whereas the technological gap is still
wide, it is narrowing at a faster speed.
3 Rocha et al. (2010a) notes the essential role that public banks fulfil in the region by providing financing
to the SMEs. The authors note that private banks are unable to fill this gap largely due to the generally
weak quality of financial infrastructure, including the availability and reliability of information on
potential borrowers.
5

Chapter 4: Impacts of bank regulations on efficiency
T
his chapter focuses on a very specific question: Are the banks in the Mediterranean more
cost efficient in countries with sounder regulatory and supervisory conditions? The
results echo the recent findings in the literature. Certain regulatory aspects, such as
disclosure requirements, credit information availability and entry obstacles, are highly
important. The presence of an explicit deposit insurance scheme also improves efficiency,
drawing attention to the importance of enhancing confidence for depositors. Other findings are
less clear and require further investigation. For example, although restrictions on activities
lower efficiency, it is possible that they could lead to increased risks.
The results of this chapter clearly show that the banks in countries with a sound
regulatory structure are significantly more efficient. In particular, the presence of deposit
insurance schemes, disclosure standards that facilitate private monitoring and the availability of
and access to credit information all enhance the cost efficiencies of banks. The stringency of
capital requirements also has a positive impact on bank efficiency, albeit to a less extent. In turn,
according to our findings, there is a case for allowing banks to engage in a wider scope of
activities and dismantling entry obstacles. Supervisory independence seems to have no impact
on cost efficiency, most likely due to the offsetting impact of increased risks arising from
concentrated political power.
In short, our results support mainly the third pillar of Basel II with weaker support for the
capital requirements. The rapid deployment of private credit bureaus, possibly modelled after
the regional best-practice as evidenced by Morocco’s brand new system, is also important in
enhancing efficiency. However, none of these factors should be treated in a vacuum.
Institutional quality, measured here by an aggregation of a number of political and governancerelated
factors, is a substantially important factor. Lastly, macroeconomic stability is also an
important contributor to the efficiencies of banks.
It should be highlighted that the results of this chapter have assessed the importance of
regulatory and supervisory practices for achieving bank efficiency. Other issues should also be
considered for making a broad assessment of the suitability and adequacy of certain rules and
standards. For example, while certain regulatory conditions may improve banks’ cost
efficiencies, they may undermine profits (e.g. systemic stability).
Chapter 5: Impacts of bank regulations on growth
This chapter turns to a broader investigation of the economic benefits of regulatory and
supervisory practices. The main question is whether banking regulations and practices
have an impact on growth. Several channels through which the relationship may operate
are considered, including the impact of regulations on cost efficiency, issuance of credit to the
private sector and capital market activity. The empirical analysis also controls for the presence
of other intermediate channels that are not accounted for.
According to our specifications, financial regulations have a relatively limited direct
impact on growth. Among the seven regulatory areas considered throughout the paper, only
government ownership—a proxy for entry obstacles and market conditions—appears to have a
consistent and significant negative impact on growth. Moreover, there is limited evidence that
financial development leads to economic development. Although efficiency and stock market
turnover appear to increase income per capita growth, private credit appears to have little (and
possibly negative) impact.
6

These results show that regulatory factors operate mostly through the financial variables.
Based on the results reviewed in this section, the regulatory factors considered in the paper
have at best an indirect impact on growth, working their way through financial development.
Moreover, lack of corruption has a clear impact on growth, which underlines its importance as a
precondition for growth.
Several technical shortcomings have to be noted at this stage. First, due to the small
sample size considered in the study, panel estimations were not feasible. Second, the similarities
between the countries considered might have generated sampling biases, which imply that the
results have to be interpreted with care and should be adequately re-assessed before applying
to other regions. Third, the similarities between countries also make the task of finding strong
indicators more difficult as the cross-country variation is relatively limited. This is indeed one of
the main causes for the apparent weaknesses of the instruments for the share of private credit in
GDP. Lastly, the regulatory variables are assumed to remain fixed over long periods; although
this assumption is unlikely to lead to substantial biases, it creates another source of
homogeneity.
Conclusions
This study sheds light on the changing regulatory environments of four south
Mediterranean countries: Algeria, Egypt, Morocco and Tunisia. Over the past two
decades, all four countries have engaged in financial sector reforms, with varying degrees
of depth, engagement and success. Morocco has achieved the most advanced financial system
as compared to the three others, eliminating interest rate subsidies and controls; reinforcing the
responsibilities and roles of the supervisor; improving the risk-management practices along
with the state-of-the-art; successfully implementing of a deposit insurance scheme; and
introducing a credit information system that may well serve as a best-practice for other
developing financial systems.
The other South-MED countries examined in this study have been less successful in
implementing key reforms. The banks in all of the three countries have relatively poor asset
qualities, as evidenced by high rates of non-performing loans (NPLs). The policies put in place
to respond to low asset quality have either led to limited improvement, a decline of credit
availability or both. The privatisation efforts have been only partly successful and at times have
not led to any change in the market conditions and financial development. In Algeria, the
publicly-owned banks continue to dominate the banking sector, accounting for over 90% of total
assets. In Egypt, although privatisation efforts have been partly successful, public loans and
debt represent a substantial proportion of the portfolios of banks, which hampers financial
development and growth opportunities. In Tunisia, a majority of the top three banks remain
owned by the state.
The comparisons among the EU-MED countries reveal particular shortcomings. Despite
some recent improvements, entry obstacles continue to be widespread in all of the South-MED
countries, arising from high rates of denied foreign applications and closely linked with a
dominant state ownership. Capital requirements are less stringent in the Southern
Mediterranean under examination, increasingly so due to the disparities in the risk-sensitivity
of the minimum capital requirements. The existing deposit insurance schemes in place, i.e. those
in Algeria and Morocco, provide adverse incentives and may increase moral hazard risks. In
Egypt and Tunisia, the implicit government guarantees may also aggravate the moral hazard
problem. Although private monitoring and disclosure requirements appear in line with the EU-
MED standards, accounting practices are increasingly poor in the South-MED. Lastly, despite
7

ecent improvements, especially in Morocco and Egypt, credit information availability is
relatively low within the region.
Turning to the cost efficiency analysis, results indicate an overall improvement in
efficiency levels for the EU-MED and South-MED in the later stages of the analysis, from 2005
onwards (with the exception of Egypt). For the South-MED, this improvement is particularly
remarkable for Moroccan and Algerian banks, but for different reasons. The overall mean
efficiency in the region is improving, once more driven by improvements in the best practice.
EU-MED banks, in particular the Spanish banks, dominate the region, with average efficiency
scores of 80.4% against the region's average of 63.5%. Spanish banks also exhibit the highest
meta-technology ratios and the ratios increase over time. This indicates that Spanish banks
consistently improved their performance, and their banking technology became best practice.
Nonetheless, during this period of analysis, the average meta-technology ratio is increasing,
which indicates an ability of banks in all countries to appropriate the best available technology.
These results are supported by the estimation of β-convergence. The β coefficient is always
negative and statistically significant, thus indicating that convergence in efficiency scores has
occurred across countries in the MED-11 area. Furthermore, results for the σ-convergence
suggest an increase in the speed of convergence as the σ coefficient is always negative and
statistically significant. This indicates that, whereas the technological gap is still wide, the gap is
narrowing at a faster speed.
When examining the impact of the regulatory and supervisory practices on cost efficiency
of banks, the results clearly show that a sound regulatory structure is a forceful contributor to
an efficient system. The case of Morocco is revealing in this respect. In particular, deposit
insurance schemes, adequate disclosure requirements and credit information availability seem
to improve the efficiencies of banks. A broader definition of the banking market by imposing
fewer scope restrictions and removing entry obstacles also improves efficiency, albeit less
significantly so than the previous factors. The rapid deployment of private credit bureaus,
possibly modelled after the regional best-practice as evidenced by Morocco’s brand new
system, is also important in enhancing efficiency.
Lastly, the pro-growth impact of regulatory adequacy appears to operate mainly though
its impact on financial development. The study shows that government ownership in banking is
detrimental to growth, even outside the scope of financial development and other financerelated
variables. Moreover, more restrictive disclosure and capital requirements as well as less
limited scope restrictions have pro-growth impacts by enhancing financial development.
It is important to note that the regulatory practices and adequacy factors should not be
treated in a vacuum. Institutional quality, measured in the study by a variety of political and
governance-related factors, is a substantially important factor in all of the regressions. The
control of corruption and the presence of democratic institutions are also important factors,
which need to be considered alongside the regulatory conditions.
To sum up, the study highlights some of the key shortcomings of the banking regulations
of the South-MED countries. It appears that some of the newer standards, such as the Basel II
capital requirements, have been conceived with developed nations in mind and may not be
appropriate, due to a variety of deficiencies in information-sharing and institutional and
disclosure mechanisms. A key aim of the upcoming reforms should be to look for ways to
reduce the role of government in the banking sector while ensuring that the regulatory
framework and the relevant institutional development adequately respond to the market
imperfections.
8

CONVERGENCE OF BANK REGULATIONS
ON INTERNATIONAL NORMS
IN THE SOUTHERN MEDITERRANEAN
IMPACT ON BANK PERFORMANCE AND GROWTH
RYM AYADI
EMRAH ARBAK
SAMI BEN NACEUR
BARBARA CASU
WITH CONTRIBUTIONS FROM
MOHAMMED YAZID BOUMGHAR
JAWAD KERDOUDI
MOEZ LABIDI
CENTRE FOR EUROPEAN POLICY STUDIES (CEPS)
BRUSSELS

TABLE OF CONTENTS
Introduction.................................................................................................1
1. Overview of the National Banking Sectors in North Africa.........3
1.1 Introduction ................................................................................................3
1.2 Algeria .........................................................................................................5
1.3 Egypt..........................................................................................................11
1.4 Morocco.....................................................................................................18
1.5 Tunisia .......................................................................................................26
1.6 Summary ...................................................................................................32
2. Convergence of Banking Sector Regulations.................................34
2.1 Methodology ............................................................................................34
2.2 Area I: Scope Restrictions .......................................................................38
2.3 Area II: Entry obstacles ...........................................................................40
2.4 Area III: Capital requirement stringency .............................................45
2.5 Area IV: Supervisory authority..............................................................49
2.6 Area V: Deposit insurance......................................................................53
2.7 Area VI: Private monitoring...................................................................58
2.8 Area VII: Credit information and laws.................................................60
2.9 Conclusions...............................................................................................65
3. Analysis of Efficiency and Convergence ........................................70
3.1 Literature review......................................................................................71
3.2 Methodology ............................................................................................74
3.2.1 Data Envelopment Analysis......................................................74
3.2.2 Meta-frontier analysis.................................................................75
3.2.3 Modelling convergence..............................................................76
3.3 Descriptive statistics ................................................................................77
3.3.1 Input and output variables........................................................79
3.4 Results........................................................................................................82

3.4.1 Efficiency results ........................................................................ 82
3.4.2 Convergence results................................................................... 88
3.5 Conclusions.............................................................................................. 90
4. Impact of Bank Regulations on Efficiency..................................... 91
4.1 Literature review..................................................................................... 91
4.2 Methodology and data ........................................................................... 93
4.2.1 Bank-specific variables .............................................................. 95
4.2.2 Country-specific variables ........................................................ 96
4.2.3 Regulatory variables .................................................................. 97
4.2.4 Data sources .............................................................................. 100
4.3 Empirical results.................................................................................... 102
4.4 Conclusions............................................................................................ 110
5. Impact of Bank Regulations on Growth....................................... 111
5.1 Literature review................................................................................... 112
5.2 Methodology and data ......................................................................... 115
5.3 Results..................................................................................................... 124
5.4 Conclusions............................................................................................ 129
6. Conclusions ........................................................................................ 130
References ............................................................................................... 133
Annex 1. CEPS Survey on Banking Supervision for the Southern
Mediterranean Countries................................................................. 145

INTRODUCTION
I
international standards and norms on banking regulations have, once
again, leaped to the forefront of the policy discussion in developed
nations due to the recent crisis in the world’s financial markets. This
discussion is far from new, nor does it apply exclusively to the world’s
most advanced economies. A sound and well-enforced regulatory regime
can help developing nations to channel financial resources more efficiently
into investments. For open economies, it can also act as a buffer, an
important stability factor in today’s shaky market situation.
Against this backdrop, this study examines the impact of banking
sector regulations on bank efficiency and economic growth in four
Southern Mediterranean countries (referred to collectively as “South-
MED”) – Algeria, Egypt, Morocco and Tunisia – while exploring the level
of convergence of regulatory practices and efficiency to EU Mediterranean 1
standards.
In particular, the study first compares the banking sector and its
regulations in Algeria, Egypt, Morocco and Tunisia, to international
standards using measures on the adequacy of regulatory and supervisory
practices. Second, banking efficiency and the level of convergence to best
practices are examined using Data Envelopment Analysis (DEA) and
complemented by Meta Frontier Analysis and the β-convergence and σconvergence
methodologies. Third, the impact of the regulatory
environment on the efficiency of banks is investigated using the developed
measures of regulatory and supervisory practices. In addition to the
regulatory details, the performance analysis also considers the legal and
institutional characteristics of the South-MED countries. Fourth, the study
1 The European Mediterranean countries considered in the study are Cyprus,
Spain, Greece, Italy, Malta and Portugal (referred to as “EU-MED”).
| 1

2 | INTRODUCTION
explores how compliance with these standards and norms may influence
the growth potential of each country.
Chapter 1 provides a descriptive analysis of the banking sectors of the
South-MED countries covered in the study. Chapter 2 then develops
measures of regulatory adequacy in a number of areas and provides
comparisons with the EU-MED. Chapter 3 summarises the analysis of
efficiency and convergence between the South-MED and EU-MED. Chapter
4 analyses empirically the determinants of the efficiency scores, paying
special attention to the regulatory adequacy measures developed. Chapter
5 provides a similar analysis for economic growth. The final chapter
concludes and puts forth the main policy recommendations.

1. OVERVIEW OF THE NATIONAL
BANKING SECTORS IN NORTH AFRICA
1.1 Introduction
A well-functioning financial system is instrumental in attaining balanced
and sustainable development. Such a system increases the availability of
funding by mobilising idle savings, facilitating transactions and attracting
foreign investments. It can also improve the allocation of financial
resources by enhancing risk management, transparency and corporate
governance practices, reinforcing property and creditor rights. Developed
financial systems are crucial in providing funding to more opaque
borrowers, such as the first-time and low-income borrowers or small and
medium-sized enterprises (SMEs), which represent a significant proportion
of economic activity but often lack the internal sources to grow. 2 In short,
financial development can serve to ameliorate the distribution of
opportunities and improve income equality. 3
The emerging consensus in the academic literature is that financial
development is possible as long as certain conditions are present to ensure
that financial intermediaries serve the financial needs of the citizens and
the private sector. These conditions include an adequate and operational
regulatory structure, a well-defined supervisory authority, legal systems
that reinforce property and creditor rights, restrained control of
government over the financial system and macroeconomic stability.
2 See Levine (1997 and 2004) and Demirgüç-Kunt & Levine (2008) for a review of
the literature on financial development and growth.
3 Although alternative theories exist, financial development is often thought to
improve income equality by enabling a more just capital accumulation or skills
development (Banerjee & Newman, 1993; Galor & Zeira, 1993). See Liang (2006) for
empirical support and a recent review of the relevant literature.
| 3

4 | OVERVIEW OF THE BANKING SECTORS
Although the four countries covered in this study have embarked
upon wide-ranging financial reforms, financial development has remained
relatively limited. With mainly bank-driven financial systems, the surveyed
countries are ‘under-banked’, i.e. the availability of credit to households
and businesses is limited. In Algeria and Egypt, these shortcomings are
quite severe, with bank loans to private enterprises representing less than
one-half of the country’s GDP (see Figure 1.1).
Figure 1.1 Private credit in the Mediterranean and Middle East (% of GDP, 2008)
Source: Beck & Demirgüç-Kunt (2009).
This section provides a descriptive analysis of the financial systems of
the four South-MED countries. The role of the public sector and the
changing regulatory and legal framework are assessed qualitatively to
highlight the main shortcomings of the banking system. Apart from legal
sources and international assessments, the discussion relies on quantitative
measures of the banking sector in each country, ranging from structure of
banking, details on aggregate balance sheets and indicators of financial
soundness.

1.2 Algeria
CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 5
Algeria’s banking system is characterised by an exceptionally strong and
persistent presence of the public sector. The public banks direct the
country’s vast domestic savings to the state-owned enterprises operating in
the country’s hydrocarbon sector, which produces the country’s chief
exports. The Algerian government has expressed interest in liberalising the
sector, although these promises are not backed by realistic policies to
attract foreign investment. Moreover, although the banks appear to be wellcapitalised,
the loan quality is very low, especially in the portfolios of
public banks, requiring constant restructuring. Access to banking services
is limited, with over 25,000 inhabitants per branch, or more than twice the
regional average (Table 1.1).
Table 1.1 Structure of the Algerian economy and banking system
2005 2006 2007 2008 2009
GDP per capita ($) 3,137 3,470 3,904 4,940 4,027
GDP per capita growth (%) 3.37 -0.70 1.20 2.22 0.52
Inflation (%) 1.64 2.33 3.56 4.86 5.74
Deposit rate 1.94 1.75 1.75 1.75 1.75
Lending rate 8.00 8.00 8.00 8.00 8.00
Commercial bank assets
(% of GDP)
56 61 69 66 72
Top-3 banks (% share) 48 54 56 57 57
Number of commercial banks 19 18 19 21 21
… of which: public 6 6 6 6 6
Number of branches 1,183 1,227 1,278 1,287 1,301
… of which: public 1,063 1,097 1,126 1,093 1,057
Inhabitants per branch 27,772 27,181 26,489 26,699 26,772
Sources: IMF and Bank of Algeria.
Prior to the 1990s, as in most centrally planned countries, Algeria had
a financial system that could only be described as “financially repressed”.
A set of regulations, laws and other non-market restrictions prevented the
intermediaries from functioning at their full capacities. The government
had full administrative control over the banking sector as a whole, setting
credit and deposit interest rates, directing the allocation of credit and
having ownership stakes in practically all credit institutions. The banking
system had the single aim of providing liquidity for the execution of the
objectives of the plan. The directed allocation of credit and high liquidity
levels led to an unrestrained monetary expansion.

6 | OVERVIEW OF THE BANKING SECTORS
The sector was partially liberalised in 1990 with the entry into force of
the Monetary and Credit Law (Law No. 90-10). The law was designed as a
first step to replace the state’s direct control over the banking sector,
facilitating entry and liberalising interest rates, first the deposit rate and
then the credit rate in 1994. Between the years 1995-2007, several private
banks were formed, almost all foreign-owned subsidiaries of prominent
international banks. 4 Although the government has repeatedly renewed its
commitment to liberalising the sector, the private banks’ role remains
limited in terms of the mobilisation and allocation of resources, except for
loans to the private sector where public banks’ predominance is lower
(Table 1.2).
Several reasons may explain the deficient growth of the privately
owned banks in Algeria. First and foremost, the public authorities have
been historically hesitant to open the banking sector to competition, since
this may divert financing from public entities. As an example, Algeria’s
ability to attract foreign direct investment (FDI) flows has been severely
hampered by the recent Supplementary Budget Law of 2009 (Ordinance
No. 09-01, Art. 58). This law requires—among other things—that the
majority stake must belong to a domestic partner for all new incoming FDI
flows. 5 This new requirement will continue to hamper the privatisation of
public banks, which puts in question the government’s willingness to
liberalise the market.
Second, Algeria has traditionally been seen as a risky country, mainly
due to political risks. 6 The recent financial crisis appears to have reduced
the investors’ risk appetite, which is often cited as the main reason behind
4 The most notable private banks include Calyon-­‐Algeria (France), Société Générale
(France), BNP Paribas El Djazair (France), Natixis (France), Gulf Bank Algeria
(Kuwait), Arab Banking Corporation (Bahrain), Citibank (US), HSBC (UK), Al
Baraka Bank (Bahrain) and Fransabank El Djazair (Lebanon).
5 Other elements of Ordinance No. 09-01, Art. 58 that relate to the FDI inflows
include i) a requirement to obtain approval from the National Investment
Development Agency (ANDI) for all new investments, and ii) the rights granted to
the state and public enterprises to have a “pre-emptive purchase right for all sales
by or to foreign investors” (see IMF, 2010a).
6 According to Euromoney magazine’s recent bi-annual survey, Algeria ranked in
101 st place worldwide among 185 countries in September 2010, the lowest among
the four Southern Mediterranean countries.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 7
the cancellation of privatisation of Crédit Populaire d’Algérie (CPA), one of
the largest public banks in Algeria.
Table 1.2 Market shares of Algerian public banks
2005 2006 2007 2008 2009
Total assets 91.4% 91.7% 92.2% 90.8% ..
Private sector loans 85.4% 83.3% 79.4% 77.0% 76.7%
Public sector loans 99.9% 99.9% 99.8% 99.8% 99.9%
Total deposits 93.3% 92.9% 93.1% 92.2% ..
Branches 89.4% 88.1% 84.9% 81.2% ..
Source: Bank of Algeria.
Table 1.3 Assets and liabilities of Algerian banks (billions of Algerian Dinars)
2004 2005 2006 2007 2008 2009 2010*
Assets 3,893 4,210 5,229 6,511 7,287 7,327 7,510
Reserves 281 198 274 445 370 340 617
Balances with foreign
institutions
77 91 84 108 142 58 55
Balances with state 803 876 1,015 941 678 810 805
…of which: Treasury
deposits
15 14 84 46 52 83 64
… of which: Deposits at
CCP
11 4 12 7 5 15 10
… of which: Treasury
bills
669 644 818 793 491 541 552
of which: Other 109 215 102 95 130 171 180
Loans 1,534 1,779 1,904 2,204 2,614 3,085 3,185
… of which: Public
enterprises
858 882 847 989 1,202 1,485 1,570
… of which: Public
authorities
0 0 1 0 0 1 1
…of which: Private
enterprises
675 896 1,056 1,213 1,411 1,598 1,613
… of which: Private
banks
2 1 0 2 1 1 1
Other assets 1,198 1,265 1,953 2,813 3,482 3,034 2,891

8 | OVERVIEW OF THE BANKING SECTORS
Liabilities 3,893 4,210 5,229 6,511 7,287 7,327 7,510
Capital 142 152 161 171 184 302 300
Reserves 25 20 24 28 169 170 173
Liabilities to nonresidents
116 84 115 90 134 46 54
Deposits 2,705 2,961 3,517 4,322 4,938 4,732 4,880
… of which: Sight:
Public enterprises
697 774 1,164 1,832 2,056 1,427 1,511
… of which: Sight:
Private enterprises
274 321 442 563 721 904 910
… of which: Sight: Other
banks
157 129 144 166 170 173 165
… of which: Term:
Public enterprises
254 366 365 351 394 499 509
… of which: Term:
Private enterprises
1,189 1,233 1,271 1,396 1,573 1,723 1,779
… of which: Term: Other
banks
134 138 130 14 24 7 8
… of which: Central
government
67 99 144 218 400 445 444
Funds by state 49 55 34 29 16 15 14
Other liabilities 790 840 1,236 1,654 1,446 1,618 1,645
* March 2010 figures.
Source: Bank of Algeria.
The predominance of state-owned banks leads to a number of
problems. First, by providing funding primarily to the public sector, the
present structure severely restricts the diversification opportunities for the
Algerian economy. According to recent figures, the share of loans to the
private sector represent only one-fifth of total banking assets (Table 1.3). A
large proportion of the total assets are held in loans to public enterprises,
mostly in the hydro-carbons sector, and balances with the state. Although a
more effective financial intermediation and diversification of the economy
are the key aims of the authorities, progress has been limited in channelling
the domestic savings into the real economy, especially to non-hydrocarbon
businesses and private enterprises (IMF, 2010a).
A second problem arises from the allocation of credit to inefficiently
run public undertakings: in particular, the state-dominated banking sector
has been characterised by exorbitant levels of non-performing loans

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 9
(NPLs), especially for loans to public enterprises (Table 1.4). Owing to their
limited role, the public banks lack the institutional framework and
experience to promote efficient intermediation. Over the past years, this
fundamental weakness has repeatedly threatened the viability of the
quality of public banks portfolios, calling for a frequent clean-up of the
balance sheets via government loan purchases. The government
implemented such a buy-back program in 2008, when the NPL rate in
public banks had dropped from 24% of total loans in 2007 to 20%. Despite
these policies, the NPL rates continue to remain high for the publiclyowned
banks, not only for their loans to state-owned enterprises but also
for the credit they extended to private-sector businesses. In turn, foreignowned
private banks, which have almost no exposures to the public sector
businesses, have relatively low NPL rates, except in recent years due to the
financial crisis.
Table 1.4 The Algerian banking system: Key financial soundness indicators
2005 2006 2007 2008
Regulatory capital (as % of RWA) 12% 15% 13% 17%
…of which: public banks 12% 14% 12% 16%
…of which: private banks 19% 22% 18% 20%
NPLs* ( as % of total loans) 19% 18% 22% 18%
…of which: public banks; all loans 20% 19% 24% 20%
…of which: public banks;
private loans
10% 12% 19% 16%
…of which: private banks; all loans 3% 3% 9% 7%
Return on equity 8% 19% 25% 25%
…of which: public banks 6% 17% 24% 25%
…of which: private banks 5% 23% 28% 26%
* Non-performing loans (NPLs) include loans in arrears with 100% provisioning
requirement. The figures include all public and private bank, including the branches of
foreign institutions.
Source: IMF (2010a).
In recent years, Algerian authorities have launched a number of
additional initiatives aimed at increasing the banking system’s lending
capacity, increasing the minimum capital requirements for banks, and
reducing the level of non-performing loans (NPLs) through financial
restructuring of public enterprises. The 1990 Monetary and Credit Law

10 | OVERVIEW OF THE BANKING SECTORS
foresaw the gradual implementation of the capital requirements set by the
1988 Basel Capital Accord. After several revisions, the minimum capital
adequacy ratio of 8% was established in 1999 and has been maintained ever
since. Reserve requirements were first implemented in 1994 (Instruction
No. 73-94), gradually raised from a minimum of 2.5% of deposits to 8% by
2008.
A deposit insurance scheme was introduced in 1997 (Law no. 97-04),
providing a guarantee of up to 600,000 Algerian dinars (approximately
€6,200 at end-2010 conversion rates) per depositor with no co-insurance or
legally set delays for making payments. The scheme became operational
with the creation of Société de Garantie des Dépôts Bancaires (SGDB) in
2003. Under the current regulations, the scheme is funded by an annual
premium charged on each bank, which is set at 1% of deposits. 7 The scheme
was put to use for the first time in 2004, to reimburse depositors of the
now-defunct El-Khalifa Bank, which was a private bank that was founded
in 1998. As noted in the World Bank (2004) assessment, there has been
some concern that the scheme lacked functional and financial
independence, which could result in discretionary decisions on its use.
Turning to legal and informational infrastructure for getting credit,
the World Bank’s Doing Business (2010) ranking clearly shows that Algeria
lags behind others, putting it in 135 th position out of a total of 183 countries.
The finding is not surprising. Algeria is behind many countries in terms of
creditors’ rights and information-sharing capacity. In particular, there are
no private credit bureaus and the public credit registry’s coverage is largely
insufficient. Secured creditors’ ability to make claims on collateral is illdefined,
severely undermining their rights and more generally the legal
framework for credit.
To conclude, Algeria’s banking system is dominated by six public
banks, which continue to collect over 90% of the domestic deposits and
divert a significant proportion to the mostly inefficient public enterprises
concentrated in the hydrocarbon sector. Under current conditions, the
Algerian financial sector is not providing the necessary funding for its
private sector to successfully diversify its economy in the near- to midterm,
in line with the aims set in its 2009 Action Plan 8 and the EU-Algeria
7 Order No. 03-10 for Money and Credit.
8 The 2009 Action Plan is available in French at http://www.premierministre.gov.dz/images/stories/dossier/Plan_action_2009_fr.pdf.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 11
National Indicative Plan (NIP) for 2011-2013. 9 In addition to hampering
growth opportunities, directed credit undermines credit quality and real
intermediation. Algeria has ample fiscal space due to its immense
hydrocarbon receipts and can continue to engage in risky loans and their
restructuration at regular intervals. However, these strategies should be
used to resuscitate private sector growth and not to support the ailing
public enterprises. In conjunction with steps to liberalise the economy, the
government should improve the pre-conditions for easing the flow of credit
information and improve creditors’ rights. Doing otherwise may hamper
credit growth for the traditionally opaque firms, such as small- and
medium-sized enterprises (SMEs), and may prolong the realisation of the
benefits from a well-developed financial sector. The use of an explicit
deposit insurance scheme when a substantial proportion of the banking
sector is publicly-owned should also be assessed.
1.3 Egypt
Egypt’s banking regulations have undergone significant reforms in recent
years. A new banking law (Law no. 88) enacted in 2003 unified all Egyptian
banking regulations, reinforcing the independence and regulatory role of
the Central Bank of Egypt (CBE), aligning the prudential standards with
the Basel II Accord as well as strengthening loan classification rules,
remedial powers of regulator, risk-based focus on supervision and capital
requirements. More recent reforms have sought to address reducing the
state-owned stakes in joint-venture and public banks, enhancing credit
conditions and increasing access to banking.
The pervasiveness of state ownership is one of the key challenges of
the Egyptian banking system. The government-controlled banks account
for nearly two-thirds of the banking activity. This has undermined
competition by obstructing entry and contributes to inefficiencies, as is
evident from the country’s high non-performing-loans. By 2006, the
reported stock of NPLs amounted to nearly a quarter of gross loans, mainly
held in public banks’ portfolios. This has resulted in a set of programmes
initiated by the Central Bank of Egypt to settle these loans by cash
injections (funded by privatisation receipts), settlements and investment
sales. Although the private banks have been increasing their network in
9 The 2011-2013 EU-Algeria National Indicative Plan (NIP) is available at
http://ec.europa.eu/world/enp/pdf/country/2011_enpi_nip_algeria_en.pdf.

12 | OVERVIEW OF THE BANKING SECTORS
recent years, most of the branches remain in urban areas. Access to banking
remains low, as in Algeria, with over 24,000 habitants per branch (see Table
1.5).
Table 1.5 Structure of Egyptian economy and banking system
2004-05 2005-06 2006-07 2007-08 2008-09 2009-10
GDP per capita ($) 1,283 1,506 1,771 2,160 2,450 1,283
GDP per capita growth (%) 2.38 4.90 3.74 4.89 2.62 2.38
Inflation (%) 8.80 4.20 10.95 11.70 16.24 8.80
Deposit rate (%) 7.23 6.02 6.10 6.58 6.49 7.23
Lending rate (%) 13.14 12.60 12.51 12.33 11.98 13.14
Bank assets (% of GDP) 131 123 126 121 105 101
Top-3 banks (% share) 58 59 57 55 .. ..
Public banks (% of total
assets)
Number of commercial
banks
.. .. .. 67 .. ..
52 43 41 40 39 39
… of which: public 7 7 6 6 5 5
Number of branches 2,841 2,944 3,056 3,297 3,443 3,490
… of which: public 2,185 2,222 2,074 2,089 2,088 2,080
Inhabitants per branch 25,308 25,000 25,360 23,992 24,252 24,212
Note: Deposit rates correspond to interest rates for three-month time deposits.
Sources: IMF, Central Bank of Egypt and Beck & Demirgüç-Kunt (2009).
Under the reform programme initiated in 2004, a number of publiclyowned
banks with high non-performing loans have been eliminated by
sales, purchases and mergers. In 2006, the country’s fifth largest bank, Bank
of Alexandria, was successfully privatised with its acquisition by the Italian
Sanpaolo IMI group. These measures have reduced the share of the
publicly-owned banks. Nevertheless, the state continues to maintain a
significant proportion of the banking system, either directly, as is the case
of the top-two banks, which are public, representing over one-quarter of
total assets, or indirectly via partial stakes, as is the case in a number of
specialised banks and joint-venture stakes held by public banks. There are
concerns over the future roles of the remaining state banks, as the private
banks continue their growth and expansion into underserved sectors,
(World Bank, 2008).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 13
The state’s influence in banking is not limited to its direct or indirect
ownership. More broadly, the level of government debt held by banks has
been very high in Egypt, accounting for an increasingly greater share in the
banks’ balance sheets. As of October 2010, the share of public debt and
loans represents around 40% of the total assets of Egyptian banks, up from
less than a quarter in 2001 (Figure 1.2).
Figure 1.2 Assets of Egyptian banks
Source: Central Bank of Egypt.
One chief underlying reason for banks’ increasing willingness to hold
more public debt than private credit is the attractive yields offered by the
treasury bonds, crowding out credit to other segments. 10 The aggregate
information on the banks’ balance sheets gives details on the activities of
Egyptian banks (Table 1.6). More specifically, public debt held by the banks
10 Since 2004, the Egyptian Treasury bills have often had yields exceeding 9%,
significantly above the overnight and short-term (3-month) deposit rates, which
remain around 6 to 7%. These conditions allow the Egyptian banks to earn
handsome amounts by simply collecting deposits and investing them in public
debt.

14 | OVERVIEW OF THE BANKING SECTORS
has increased substantially in recent years and has surpassed the total
outstanding private loans as of October 2010. Meanwhile, customer
deposits represent nearly two-thirds of the total balance sheet.
There is some evidence that the policies implemented since the early
2000s to diminish NPLs have also made banks more reluctant to lend to the
private sector, especially to more risky lines of business. Deficiencies in the
availability of credit-worthiness information and managerial skills of most
small- and medium-sized enterprises (SMEs) have also contributed to
greater-than-normal risks in the real sector. Outstanding bank credits are
by-and-large concentrated in the blue-chip corporations, with retail and
SME sectors remaining relatively underserved due to inherent risks.
Facilitating bank intermediation remains one of the key challenges in Egypt
and has been the main aim of the recent reform initiatives (IMF, 2010b).
In recent years, Egypt has made significant strides in the field of
credit information-sharing. The Central Bank of Egypt, which supervises
the Egyptian banking system, has operated a public credit registry (PCR)
since 1957. In 2002, an online system was adopted, permitting banks to
extract and transmit information electronically. In 2004, the coverage of the
registry was improved by lowering the threshold for reporting credit to
E£30,000 (approximately €4,000) (Emerging Markets Group, 2006). In the
same year, the CBE required the larger credit institutions to report any
credit card delinquencies (defined as 90 days past due and/or with legal
proceedings) in an effort to compile a “Negative List Database”; however,
the database is not distributed directly by the public registry. Legal
amendments introduced in 2006 allowed the public registry to share
information with the country’s first private credit bureau, I-score, founded
by a consortium of the country’s main banks in 2005. Other banks, nonbank
financial institutions and utility companies have also been invited to
join the private credit bureau in 2008. With these changes, the private
database is expected to cover a large majority of the lending portfolio of all
banks.

Table 1.6 Assets and liabilities of Egyptian banks (billions of Egyptian pounds)
2001 2002 2003 2004 2005 2006 2007 2008 2009 2010*
Assets 428.4 495.5 577.9 633.4 703.6 761.6 937.9 1,083.3 1,092.0 1,220.7
Cash 3.5 4.5 5.6 5.4 6.6 6.8 7.7 10.3 11.1 12.4
Public debt 71.1 87.7 111.3 137.4 170.7 194.0 176.1 201.9 332.6 405.9
Balances with domestic
67.0 83.2 110.9 116.3 125.0 121.7 217.4 278.2 173.5 200.7
banks
Balances with foreign banks 16.3 20.0 29.8 43.3 51.2 72.6 124.4 122.8 77.1 57.4
Loans 241.5 266.1 284.7 296.2 308.2 324.0 353.7 401.4 430.0 466.0
… of which: Public 42.3 45.5 48.2 51.6 59.3 53.6 50.9 57.8 63.6 69.2
… of which: Private 199.2 220.6 236.5 244.6 248.9 270.4 302.9 343.6 366.3 396.8
Other assets 29.0 33.9 35.7 34.8 42.0 42.5 58.6 68.8 67.7 78.2
Liabilities 428.4 495.5 577.9 633.4 703.6 761.6 937.9 1,083.3 1,092.0 1,220.7
Capital 12.0 12.5 18.2 20.3 22.9 27.1 33.0 37.6 41.6 46.6
Reserves 10.2 11.2 11.8 11.5 12.4 13.4 12.6 19.8 21.4 28.5
Provisions 31.2 35.9 40.1 44.6 49.5 55.0 53.5 62.3 69.7 70.4
Long-term loans & bonds 11.9 14.1 14.9 15.0 14.3 17.5 26.4 22.3 22.0 21.7
Liabilities to domestic banks 28.2 35.1 35.6 29.9 22.7 21.5 82.6 98.7 31.0 53.9
Obligations to foreign banks 11.5 11.8 16.2 10.3 12.3 8.8 10.0 13.3 18.2 20.3
Deposits 291.2 340.9 403.1 461.7 519.6 568.8 650.0 747.2 809.7 892.5
… of which: Public .. .. .. .. 108.7 108.2 107.1 126.3 139.9 151.9
… of which: Private .. .. .. .. 410.7 460.9 546.7 624.7 676.4 744.1
Other liabilities 32.2 34.0 38.0 40.1 49.9 49.5 69.9 82.1 78.4 86.8
* October 2010 figures.
Source: Central Bank of Egypt.
| 15

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 16
Despite significant improvements in credit information-sharing,
barely any progress has been noted in improving the legal and institutional
framework for credit. A number of notable weaknesses exist in loan
enforcement and collateral foreclosure practices, including an incomplete
definition of secured transactions that are allowable as collateral.
Administrative costs for registering land titles and mortgages remain high,
which make collateralisation difficult. Out-of-court enforcement remains
largely unavailable, except for secured claims over securities. In particular,
collection of unsecured debt or secured real estate transactions is only
possible through complex and lengthy court proceedings. The CBE has
sought to address these challenges through the creation of an NPL
Management Unit, the launch of a conciliation and arbitration mechanism,
and regulatory changes in the real estate finance law. However, these
moves are unlikely to be as efficient as a full-scale revision of the legal and
institutional framework for credit.
The introduction of an explicit deposit insurance scheme was
foreseen under the second phase of the banking sector regulation reform
initiated in 2008. Information obtained from the authorities and the CBE’s
website shows that the scheme aims to protect small depositors. Other
details on the scheme, such as the type of funding, risk-responsiveness of
premiums and potential government backing are not available. At this
moment, no timeline has been provided on when the scheme will become
operational.
Other reforms on the agenda that are currently at play include the
comprehensive implementation of the Basel II standards along with
supplementary prudential measures to limit excessive risk in the financial
sector. To that extent, there is some scope for cooperation with the EU for
capacity-building purposes. The authorities have also expressed an interest
in encouraging banks to publish more detailed information, where
cooperation opportunities also exist.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 17
Table 1.7 Egyptian banking system: Key financial soundness indicators
Regulatory capital
(as % of RWA)
NPLs
( as % of total loans)
Provisions (as % of
classified loans)
2001 2002 2003 2004 2005 2006 2007 2008 2009
10.2% 11.0% 11.1% 11.4% 13.7% 14.7% 14.8% 14.7% 15.1%
15.6% 20.2% 24.2% 23.6% 26.5% 18.2% 19.3% 14.8% 13.4%
69.4% 62.3% 57.0% 60.2% 51.0% 76.2% 74.6% 92.1% 100.4%
Return on assets 0.8% 0.5% 0.5% 0.5% 0.6% 0.8% 0.9% 0.8% 0.8%
Return on equity 13.7% 8.9% 8.9% 9.8% 10.2% 14.3% 15.6% 14.1% 13.0%
Sources: IMF, Global Financial Stability Reports, 2005-2010.
Although the Egyptian government has been actively engaged in a
variety of regulatory reforms in recent years, some of the endemic
problems in the banking sector continue to exist and may well remain
unaddressed in the upcoming years. Indeed, despite the privatisation
move, by the end of 2008, the market share of institutions that were wholly-
or majority-owned by the state remained over half of the total banking
assets (Table 1.5). Although the NPL rates have dropped considerably, they
still represent 13.4% of gross loans by the end of 2009. Additionally, a
significant amount of untapped financial liquidity continues to remain
dormant within the banking system. The aggregate loan-to-deposit ratio –
dropping to 53% in 2009 – points to a clear under-leveraging in the sector,
which has become a more acute problem in recent years, especially after the
recent set of reforms aimed at improving the banks’ balance sheets. The
remainder of the banking assets are held in safe and higher-yield
government debt, with treasury bills held representing nearly one-third of
total assets.
These points underscore the present trade-offs between the level of
public debt, restrictiveness of prudential regulations, extent of information
sharing, adequacy of corporate governance practices and credit availability.
Although the government has pursued an ambitious reform agenda in
some of these areas, a comprehensive assessment is necessary to ensure
that the post-reform conditions are consistent with the country’s long-term
development strategy of facilitating endemic growth in the private-sector.

18 | OVERVIEW OF THE BANKING SECTORS
1.4 Morocco
Morocco has one of the largest banking sectors in the Southern
Mediterranean, with the total assets of commercial banks representing over
120% of the country’s GDP. Commercial banks play a crucial role in the
country’s financial system and have increasingly developed links with
other financial intermediaries in the rapidly expanding insurance,
securities, leasing and factoring sectors. The banking system is relatively
concentrated, with the market share of the top three banks remaining
around two-thirds of the total bank assets. In 2009, the sector consisted of a
total of 13 privately-owned banks, seven of which are majority-owned by
foreign shareholders, and six publicly-owned banks. In addition to these
depositary institutions, there are six offshore banks and 12 microfinance
institutions, which are not included in the figures below.
The central bank, Bank Al-Maghrib (BAM), was created in 1959
(under its prior name, Banque du Maroc) to issue banknotes and coins,
safeguard the stability of the currency and to preserve the soundness of the
banking system. More specifically, the by-laws of BAM stipulate that the
chief role of the body is “to ensure the well-functioning of the banking
system and the implementation of the laws and regulations relating to the
surveillance and control of the activities of credit institutions and related
institutions”. 11 The head of the body, the Governor, is elected by the
Sovereign; the head of the supervisory unit within BAM is named by the
Governor of BAM, with an undefined tenure – possibly for life.
Macroeconomic conditions have improved in recent years, thanks to
increased foreign direct investment (FDI) and remittance inflows as well as
tourism receipts. Fiscal conditions have also recovered due to structural
reforms and fiscal consolidation efforts. 12 Since Morocco has a pegged
currency, fixed at a basket of currencies consisting of the euro and the US
dollar, and a partially closed capital account, the capital inflows have
contributed to increase domestic liquidity and, in parallel, banks’ liquid
assets. 13 The central bank (Bank Al-Maghrib-BAM) has increased the banks’
11 As translated from Article 9 of Law no. 76-03 on the status of Bank Al-Maghrib.
12 The fiscal conditions have deteriorated in 2008-10 due to a jump in subsidies and
lower tax revenues. These conditions are not expected to threaten the long-term
conditions (IMF, 2010c).
13 There are several outward capital controls in place in Morocco; inward controls
have been lifted in recent years. For currency transactions, exporters can deposit

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 19
required reserves and used its deposit facility regularly to absorb excess
liquidity and to keep price stability under control.
Table 1.8 Structure of the Moroccan economy and banking system
2001 2002 2003 2004 2005 2006 2007 2008 2009
GDP per capita ($) 1,308 1,385 1,688 1,905 1,967 2,142 2,427 2,827 2,865
GDP per capita
Growth (%)
.. 2.1% 5.1% 3.5% 1.7% 6.4% 1.5% 4.1% 4.3%
Inflation (%) .. 2.8% 1.2% 1.5% 1.0% 3.3% 2.0% 3.9% 1.0%
Deposit Rate .. 4.5% 3.8% 3.6% 3.5% 3.7% 3.7% 3.9% 3.8%
Lending Rate .. 13.1% 12.6% 11.5% 11.5% .. .. .. ..
Bank assets
(% of GDP)
85% 86% 86% 88% 96% 102% 117% 120% 121%
Top-3 banks
(% of total assets)
.. 51% 54% 64% 64% 64% 63% 65% 66%
Number of banks 19 18 18 17 16 16 16 18 19
of which: public 7 7 6 6 5 5 5 5 6
of which: foreign 7 5 5 5 5 5 5 7 7
Public banks
(% of total assets)
.. .. .. .. .. 29% 27% 26% 25%
Foreign banks
(% of total assets)
.. .. .. .. .. 22% 22% 21% 21%
Branches 1,805 1,878 1,948 2,033 2,223 2,447 2,748 3,138 3,538
Inhabitants per
branch
15,974 15,540 15,154 14,677 13,559 12,463 11,230 9,952 8,913
Note: Deposit rate is determined based on the 3-months TD rate and the lending rate is
determined by the maximum export credit.
Sources: IMF and Bank al-Maghrib.
The Moroccan banking sector has undergone significant changes
following the reform process of early 1990s. The process aimed to establish
a financial sector that serves the market economy, mobilising savings and
optimally allocating investments. The requirements for private banks to
hold development bank bonds were largely abolished by the banking law
up to 50% of foreign exchange receipts in the foreign exchange accounts. For
capital inflows, commercial banks may only borrow abroad to finance foreign
trade or investment transactions or for covering currency risks for customers. Also,
outward direct investments of resident firms and citizens are subject to approval.

20 | OVERVIEW OF THE BANKING SECTORS
of 1993. Interest rate subsidies and controls were completely eliminated in
the years that followed, with the exception of sight deposits and small
savings deposits, which continue to be non-remunerated. The more recent
2006 banking law has reinforced the autonomy and roles of the country’s
regulatory authority, Bank al-Maghrib (BAM), enlarged its control to the
entire banking sector, enhanced deposit insurance schemes and broadened
its supervisory authority.
Despite the excess liquidity, credit to the private sector remained flat
in the first half of the 2000s, remaining around 50% of GDP (Figure 1.3). A
number of underlying factors can be put forward to explain the
unresponsiveness of credit conditions to the overall availability of liquidity.
The Moroccan economy is made up of a large number of small firms
operating in the informal sector with opaque information on
creditworthiness. Additionally, the banks had a large portfolio of nonperforming
loans (NPLs), which undermined their appetite for risk. Lastly,
the handsome interest earnings from holding excess reserves also
competed with any real lending activity that the banks might undertake.
Figure 1.3 Private sector credit in Morocco (as % of GDP)
Source: Beck & Demirgüç-Kunt (2009).
Private credit growth picked up substantially in the second half of the
decade, representing over 80% of the GDP by the end of 2009. These
developments are partly explained by dropping NPL ratios, which might
have contributed to an increasing risk appetite for banks (Table 1.9).
Moreover, BAM encouraged lending by strengthening credit information
standards and risk management capacity of banks, most notably by setting

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 21
up a credit bureau in 2009. The resources of public credit guarantee
schemes have also been expanded substantially, reaching $370 million
(0.4% of GDP) in outstanding guarantees in 2009, compared to $251 million
in 2007. 14 Lastly, the rising asset prices, notably in the real estate markets
(reaching 20% in some cities in 2007), have created a wealth effect and
increased collateral values, lifting both the demand and supply of credit
(Allain & Oulidi, 2009).
Table 1.9 Moroccan banking system: Key financial soundness indicators, 2001-09
Regulatory
capital
(as % of RWA)
NPLs (as % of
total loans)
Provisions
(as % of NPL)
2001 02 03 04 05 06 07 08 09
12.6 12.2 9.3 10.5 11.5 12.3 10.6 11.2 11.7
17 17 18 19 16 11 8 6 6
53 55 55 59 67 71 75 75 74
Return on equity 10.2 1.9 -2.1 10.9 6.3 17.4 20.6 16.7 17.0
Return on assets 0.9 0.3 -0.2 0.8 0.5 1.3 1.5 1.2 1.3
Source: Bank al-Maghrib.
The aggregate balance sheet for the Moroccan banking sector
highlights some of the issues discussed (Table 1.10). Unlike their
counterparts in Algeria and Egypt, public debt represents a small
proportion of the portfolios of Moroccan banks. This is especially the case
since 2008, when the ratio of public debt and public loans to total assets
dipped below 10%.
14 It is not entirely clear to what extent Morocco’s credit guarantee scheme serves
smaller firms. According to 2009 figures, the average value of guarantees currently
stand at $155,000, or 60 times per capita income, which are much larger than
regional and global averages (Saadani et al., 2010).

22 | OVERVIEW OF THE BANKING SECTORS
Table 1.10 Assets and liabilities of Moroccan banks (millions of Moroccan Dirham)
2003 2004 2005 2006 2007 2008 2009 2010*
Assets 409,576 442,487 507,702 591,284 720,313 828,100 888,566 925,571
Balances with non-residents 15,862 19,693 28,038 30,618 32,299 28,856 28,599 23,602
Balances with central bank 38,637 47,753 48,987 53,206 65,932 61,097 42,829 32,184
Loans 232,305 249,594 289,345 352,198 455,729 561,907 630,504 680,125
…of which: Government 4,451 4,984 6,723 6,549 6,177 12,252 16,108 16,175
… of which: Non-bank private 224,369 241,513 271,983 329,483 429,208 534,061 589,555 637,144
Negotiable debt securities 88,811 85,437 99,278 101,240 103,610 101,275 97,041 98,311
… of which: Government 78,568 73,752 83,558 88,838 89,344 78,875 69,990 68,906
… of which: Non-Bank Private 2,430 3,513 6,545 7,130 6,329 12,726 14,795 14,446
Money market funds 0 0 548 4,084 1,280 1,316 1,756 2,634
Shares and other equity 13,435 18,084 16,246 21,328 25,161 35,404 45,977 45,407
Fixed assets 11,494 12,736 13,485 13,727 14,795 15,770 16,806 17,853
Other assets 9,033 9,190 11,775 14,883 21,507 22,475 25,054 25,456
Liabilities 409,576 442,487 507,702 591,284 720,313 828,100 888,566 925,571
Liabilities to non-residents 13,785 11,430 12,022 12,847 12,372 13,472 12,282 19,290
Liabilities to public & fin. inst. 17,685 17,042 28,466 36,816 48,918 57,887 74,031 84,917
Non-bank private deposits 279,474 302,863 347,158 403,273 474,915 524,672 542,820 549,676
… of which: Sight deposits 151,868 170,767 199,490 233,667 291,721 306,617 323,302 326,329
… of which: Term & saving accounts 127,607 132,096 147,668 169,606 183,194 218,055 219,518 223,348
Money market funds 0 0 15,323 22,107 30,546 39,202 50,893 59,899
Negotiable debt securities 9,592 7,559 5,672 6,873 12,236 29,223 36,031 35,956
Shares and other equity 36,271 37,074 38,986 47,137 57,041 66,266 73,212 78,966
Other liabilities 34,263 42,256 41,824 37,006 45,865 38,766 39,553 40,299
* October 2010 figures.
Source: Bank al-Maghrib.

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 23
In fact, the total outstanding public debt has declined over the last 8
years while the total assets have increased by more than two-fold. On the
liability-side, the funding is mostly through customer deposits, which have
represented roughly 60% of the total liabilities. The banks increasingly use
the money market for their funding, although their share in total liabilities
remains small.
The 2000s have also witnessed the opening of a number of
microfinance institutions. The 1999 microcredit law allowed these
institutions to borrow funds from the domestic financial market and offer
credit without being restricted by rate caps. 15 Within several years,
Morocco became a regional leader in the microfinance sector, currently
supervised by BAM (after the introduction of the new banking law in 2006).
By 2008, the Moroccan microfinance sector provided funding to over 1.2
million active borrowers and a total loan portfolio of over $700 million,
representing approximately 1% of loans to the private sector. 16
The government’s limited role in the banking sector is another aspect
that sets Morocco apart from its neighbours. This has not always been the
case as the Moroccan government maintained a substantial proportion of
banking under its control in late 1990s and early 2000s. In recent years,
however, considerable progress has been made in restructuring public
banks, sale of public shares and the full compliance of remaining public
banks with regulatory requirements by 2007. The state-owned banks
continue to represent a significant proportion of total activities, but their
market shares have declined substantially, down to 25% of total assets in
2009 from 40% in 2002 (Table 1.8).
The government’s involvement is not strictly restricted to its direct
control over the banking sector. The Moroccan government, like its
neighbours, has used the domestic banking sector to fund the public
budget. In the 1980s and 1990s, all banks were required to hold a
substantial proportion of their portfolio in treasury bonds. The banks were
also required to hold bonds issued by the various development banks,
15 Currently, microcredit institutions are not allowed to take retail deposits.
16 Partly owing to their rapid expansion and diminishing asset quality in recent
years, the Moroccan microcredit institutions have been hit hard by the 2007-09
financial crisis, facing unprecedented levels of non-performing and problem loans.
In consequence, the sector has shrunk by 6% in 2008 and is likely to face
consolidation in the upcoming years (MIX, 2009).

24 | OVERVIEW OF THE BANKING SECTORS
which were publicly owned. By late 1990s, most of these requirements were
dropped. As a consequence, the government securities now account for a
much smaller proportion of the banks’ balance sheets, dropping from 20%
of total assets in 2003 to about 7% in October 2010 (Table 1.10).
As noted in the IMF’s (2008) revised assessment, Morocco’s banking
supervision complies with the majority of the Basel Core Principles for
Effective Banking Supervision (BCP). 17 Morocco has required all banks to
apply the standardised approach to risk under Basel II since 2007, earlier
than all other countries in the region. 18 BAM has published several
guidelines for the implementation of the second and third pillars of Basel II,
in line with the Basel Committee’s recommendations. Minimum capital
adequacy levels have been shifted to 10% of risk-weighted assets (RWA) in
2008 with an intention to raise them further to 12% in the upcoming years.
Morocco is also a leader in other areas of regulation in the region. The
country is one of the two South-MED countries covered in this study (apart
from Algeria) to have an explicit deposit insurance scheme. Created in
2006, the scheme is funded by the banks and compensates depositors for
lost funds up to 80,000 Moroccan Dirhams (DH) (approximately €7,200 as
of end-2010) per depositor. 19 If the fund is insufficient to pay out all eligible
deposits, proportional haircuts are applied to the legal protection. The
funds may also be used to provide emergency credit to problem banks,
which has been identified as a potential source of conflict by the IMF in its
most recent FSAP update, even though the fund has never been used for
that purpose in practice (IMF, 2008).
As another regional ‘first’, a private credit bureau became operational
in October 2009. The bureau is developed and operated by Experian, which
is a global leader in credit information services. The setup of private credit
bureaus was heralded by a series of regulatory arrangements in 2007,
delegating the credit information exchange functions to the private sector
and effectively abolishing the similar functions of the public credit
17 In the IMF’s (2008) assessment, the country’s regulatory structure was compliant
or largely compliant with 21 of the 25 BCP principles.
18 The tier 1 requirements have been defined by Regulation No. 24/G/2006 on the
prudential capital requirements for credit institutions on individual and
consolidated bases. Upon receiving interest from several banks on the use of
internal model-based approach to risk, BAM has started work on preparing the
guiding principles with a preliminary implementation date of 2011-12.
19 The coverage was raised from 50,000 DH by the 2006 banking law.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 25
registers. 20 According to the legal framework for private credit bureaus,
BAM acts as an intermediary in the flow of information. All regulated
credit institutions, including microcredit institutions, are required by law
(i.e. mandatory reporting) to provide positive and negative information
about the creditors to BAM. The full data files are then passed over to the
credit bureau. 21
Despite the absence of formal agreements (i.e. MoUs) or participation
(i.e. observer in European Committee of European Banking Supervisors –
CEBS), the Moroccan authorities have also been relatively eager to
cooperate with their EU counterparts on select matters. Cooperation on
assisting publicly-owned institutions was in place in early 2000s. Moreover,
BAM has shown its intention to engage in cooperation on the
implementation of the internal ratings approach under Basel II and the
upcoming Basel III agreements.
To summarise, Moroccan authorities have successfully implemented
the reform programmes over the last decade to modernise the financial
services sector. Today, the country is exemplary for its banking regulations,
deposit guarantee scheme, information-sharing infrastructure and in the
microfinance sector. The main challenge the country will face in the
upcoming years will be the potential for instability from external markets
as capital flows and exchange rate policies are liberalised, as intended by
the authorities. Although the IMF’s (2008) stress tests have revealed that
the banking sector is resistant to credit, liquidity and interest risks, there
are vulnerabilities arising from concentration risks and exposure to the real
estate sector. More broadly, the authorities have to ensure that the recent
20 Regulations 27/G/2007 and 28/G/2007 of BAM. For more information on the
fundamental aspects of the private credit bureaus in Morocco, see the document
entitled “Enjeux et Modes Opératoires de la Délégation de la Centrale des Risques
de Bank Al-Magrhip”, 26 November 2007 (available at http://www.bkam.ma/
wps/wcm/resources/file/eb455a459c84942/Dpliant%20Dlgation%20de%20la%20
gestion%20du%20Service%20Central%20des%20Risques.pdf).
21 As noted in Madeddu (2010, pp. 22-23), this innovative “delegated model”
implemented by BAM is attractive as it i) prevents market segmentation through
the formation of credit bureaus that have data from only some creditors (i.e.
‘vertical informational silos’); ii) can facilitate entry by other private information
providers; iii) prevents lenders’ reluctance to share data directly with the private
bureaus; and iv) supplements the central bank’s supervisory role by creditor
information.

26 | OVERVIEW OF THE BANKING SECTORS
jump in private credit does not lead to a resurgence of non-performing
loans.
1.5 Tunisia
Following Morocco, Tunisia’s banking system is the most developed in the
region, with total assets representing nearly 97% of GDP in 2009. The
banking system dominates the financial markets, with the capital and
insurance markets representing a very small proportion of the overall
financial activities. 22 The banking sector is comprised of 18 commercial
banks, three of which remain publicly-controlled, i.e. with majority stateownership.
23 The system is relatively dispersed, with the market share of
the largest three banks accounting for about one-third of the total assets of
commercial banks.
The banking system is supervised by an organ of the central bank,
Banque Central de Tunisie (BCT), although the control over the stateowned
institutions is exercise in part by the Ministry of Finance. The head
of the supervisory body is appointed by the Governor of the central bank
with an undetermined tenure.
The macroeconomic conditions have remained relatively stable in the
past decade. Following structural reforms, fiscal and external
vulnerabilities were significantly reduced. Real GDP growth in this period
remained at around 5% while inflation remained less than 6% for most of
the decade. Despite these positive aspects, unemployment remains high, at
around 13 to 14%. The conditions are particularly dire among the youth,
with 30% of those aged 15-24 remaining unemployed. Indeed, the high
unemployment rates are blamed as one of the principal causes (in addition
to low levels of accountability) of the protests and fall of the Ben Ali
government at the beginning of 2011.
Starting with late 1990s and early 2000s, the Tunisian authorities have
embarked on an ambitious financial reform. In addition to attempts to
strengthen the credit culture, the authorities have also revised the laws on
22 The market capitalisation and turnover of the Tunis Stock Exchange remains at
15% and 4%, according to the 2009 year-end figures.
23 Other credit institutions, including development banks and microfinance
institutions, are not included in the figures and constitute a negligible proportion
of the financial system.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 27
the central bank and credit institutions in the 2000s. 24 The prudential rules,
first adopted in 1991, were revised in 2001, setting the standards on capital
requirements, reserve requirements, liquidity requirements, risk
management and relations with affiliates. 25 In the same year, new laws
were enacted to give the BCT a number of surveillance powers on
monetary and on-site supervision. 26
Table 1.11 Structure of Tunisian economy and banking system
2000 2004 2009
GDP per capita ($) 2,036 2,845 3,852
GDP per capita growth (%) .. 5.1% 1.9%
Inflation (%) .. 3.6% 3.7%
Commercial bank assets (% of GDP) 89% 89% 97%
Top-3 banks (% of total assets) .. .. 34%
Number of commercial banks 13 16 18
… of which: public 5 3 3
Public banks (% of total assets) 49% 44% 32%
Branches .. .. 1,209
Inhabitants per branch .. .. 8,541
Sources: IMF and Banque Central de Tunisie (BCT).
Although recent privatisation efforts have reduced direct stateownership,
public banks continue to play a predominant role in the
banking sector, representing just under one-third of total assets in 2009. The
three banks that remain majority-owned by the state are among the largest
four banks in the country. The largest one, which also happens to be the
second largest bank in Tunisia, Société Tunisienne de Banque (STB),
accounts for around one-third of all loans to the tourism sectors. The
second largest public bank, Banque National Agricole (BNA), provides
more than half of the loans to the agriculture and fisheries sectors. The
third public bank, Banque de l’Habitat (BH), provides nearly one-fifth of
24 Law no. 58-90 of 1958 on the creation and organisation of the Banque Centrale de
Tunisie (BCT) was amended in 2006 (Law No. 2006-26) and in 2007 (Law No. 2007-
69). The law on credit institutions (Law no. 2001-65) was amended in 2006 (Law
No. 2006-19).
25 Regulation No. 91-24 of 1991 on prudential regulations concerning banks was
revised in 2001 by Regulation 2001-04.
26 BCT’s supervisory powers are defined by Law No. 2001-65 and its amendment
Law No. 2006-19.

28 | OVERVIEW OF THE BANKING SECTORS
the real estate loans, which represent a substantial proportion of total
outstanding credit.
One of the key characteristics of Tunisia’s banking system is the
persistently low quality of assets, emanating from problem loans. In 1997,
the authorities launched a plan to tackle the problem though restructuring.
As noted in IMF (2002 and 2007), the authorities allowed banks in the early
2000s to create asset management companies as their subsidiaries in order
to purchase and pool NPLs. The problem loans to public enterprises were
similarly restructured, this time backed with government guarantees. In
2001, reporting requirements were toughened, requiring banks to obtain
detailed financial statements certified by external auditors or rating
agencies for large exposures. 27 Starting in 2004, the BCT forced banks to
allocate their net incomes and withhold dividends, if necessary, to cover
any under-provisions. Two public banks with extensive problem loans
were privatised in the same year, even though the government continues to
hold significant minority stakes. Legal reforms to facilitate recovery were
also implemented in recent years in an attempt to streamline sale of assets
and restrict undue delays in recovery of claims. More recently, the
authorities increased the provisioning requirements and expanded the tax
deductibility of provisions.
27 Detailed information requirements for large exposures were set by Regulation
No. 2001-12.

Table 1.12 Assets and liabilities of Tunisian banks (millions of Tunisian Dinar – TD)
CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 29
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010*
Assets 23,745 26,278 27,015 28,185 31,138 33,954 36,470 41,377 46,682 51,892 53,840
Cash 146 144 139 139 138 136 201 241 235 259 276
Deposits at BCT 302 605 530 558 764 1,058 1,341 1,920 3,204 3,689 2,824
Foreign assets 928 808 957 853 906 1,281 1,393 1,996 1,755 2,160 2,258
Claims on state 1,620 1,487 1,559 1,664 2,145 2,271 2,609 2,817 2,501 3,060 3,014
Private credit 14,538 16,241 17,122 18,141 19,981 21,561 23,149 25,465 29,322 32,191 34,715
Securities 746 797 1,020 1,120 1,232 1,415 1,539 1,650 1,803 2,128 2,252
Other assets 5,465 6,195 5,688 5,711 5,972 6,232 6,237 7,287 7,863 8,404 8,501
Liabilities 23,745 26,278 27,015 28,185 31,138 33,954 36,470 41,377 46,682 51,892 53,840
Sight deposits 3,583 3,959 3,697 3,919 4,265 4,721 5,422 6,271 7,000 8,263 8,795
Other deposits 8,365 9,293 10,119 10,868 12,151 13,273 14,674 16,539 19,278 21,427 22,261
Foreign liabilities 2,783 2,886 3,280 3,180 3,695 4,194 4,331 4,899 5,147 5,819 5,804
Liab. to BCT 454 870 504 444 93 4 123 17 18 2 53
Special resources 849 945 1,080 993 1,033 1,105 1,135 1,092 1,139 1,163 1,177
Equity 2,841 2,881 3,076 3,431 4,014 4,486 4,928 5,471 6,258 7,064 7,225
Other liabilities 4,870 5,444 5,260 5,350 5,888 6,170 5,857 7,089 7,842 8,153 8,525
* May 2010 figures.
Source: Banque Central de Tunisie (BCT).

30 | OVERVIEW OF THE BANKING SECTORS
Despite regulators’ attempts and a generally good performance of the
Tunisian economy in recent years, the NPLs have remained relatively high.
As shown in Table 1.13, over the last decade the ratio of NPLs to gross
loans has remained around 15 to 20% for most years in both publiclyowned
and privately-owned banks. The NPL ratios have declined in recent
years in line with the objectives set by authorities to reduce them below
15% by the end of 2009 (see Table 1.13). These developments are, at least in
part, due to the improved risk assessment practices on the part of banks
through the increased availability of borrower information, reforms to
facilitate the sale of collateral, privatisation of public banks as well as the
new tax and regulatory arrangements on provisioning. Nevertheless, the
NPLs continue to be a problem in the country’s banking sector,
representing over 13% of gross loans.
Table 1.13 Tunisian banking system, financial soundness indicators (%)
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
Reg. capital
(as % of RWA)
11.3 11.1 10.2 9.3 11.6 12.4 11.8 11.6 11.7 12.4
Private banks 10.6 10.5 10.3 8.4 12.4 13.5 12.1 9.7 11.0 11.6
Public banks 11.8 11.8 10.1 10.8 10.1 10.0 9.3 9.9 9.6 10.9
NPLs* ( as %
of total loans)
21.6 19.2 20.9 24.2 23.6 20.9 19.3 17.6 15.5 13.2
Private banks 15.4 16.1 18.1 21.6 20.4 20.0 19.0 18.1 15.3 12.5
Public banks 26.8 22.8 24.3 26.7 27.4 22.1 19.7 17.3 15.9 14.1
Provisions
(as % of NPL)
49.2 47.4 43.9 44.1 45.1 46.8 49.0 53.2 56.8 58.3
Private banks 54.7 47.7 44.9 39.9 43.5 45.9 48.4 52.0 55.0 59.2
Public banks 46.6 47.1 42.9 46.2 47.6 49.1 50.2 55.0 58.1 57.1
RoA 1.3 1.1 0.7 0.5 0.5 0.6 0.7 0.9 1.0 1.0
RoE 14.5 13.2 7.6 4.6 4.8 5.9 7.0 10.1 11.2 11.7
Sources: IMF (2007 and 2010d).
Aside from politically-connected lending, several reasons can be put
forward to explain the persistence of problem loans in Tunisia’s banking
system.
First, some of the recent jumps in problem loans in the existing loan
portfolio can be explained by the external economic environment and
events. For example, the global slowdown and the recession in the tourism
sector following the September 11, 2001 events and the April 2002 terrorist
attack in Djerba have severely affected the asset quality of the banking
system. This pushed the NPL ratios up by 4 to 5% between 2001 and 2003.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 31
Second, NPL ratios often have a lagged policy response when the
stock of older NPLs does not improve significantly over time. According to
IMF (2009), although the new NPLs remain low, the asset quality of the
existing stock of loans has not improved over the last few years. This was
largely due to the perverse incentives provided by the prior restructuring
efforts, giving banks no incentives to opt for a deeper restructuring (i.e. full
write-offs) or a thorough assessment of the debt-repayment capacity of
borrowers.
Third and last, most of the recent policies to address credit quality
problems are backward-looking and do not have a direct impact on
reducing NPLs before they arise. This includes a majority of the regulatory
arsenal put forth by the authorities to address the high level of NPLs, such
as the use of provisions or the legal reforms, which have only an ex-post
impact. Indeed, these measures can only mitigate the transaction costs and
legal uncertainties once the loans are deemed problematic. More forwardlooking
risk assessment measures are needed to minimise NPLs before they
arise. These include the adoption of Basel II standards, more regular stresstesting,
developing CAMEL-type 28 regulatory assessment tools and
enhancing the credit information environment by developing a private
credit bureau (IMF, 2010d). 29
The (previous) presidential programme of 2010-14 identifies the
strengthening of the financial sector as a key policy objective. Perhaps most
importantly, the authorities have shown their willingness to implement the
Basel II framework, although no clear timeline has been set for the adoption
of the international standards. A deposit insurance scheme is also under
preparation under the programme. Other aims include the consolidation of
the banking institutions, increasing banks’ impact on the economy,
restructuring the public sector and promoting Tunisia as a regional
banking services hub. The relevance of the programme in the aftermath of
the events of January 2011 remains to be seen.
Despite the absence of formal arrangements with EU supervisors, the
BCT will take part in a ‘twinning project’ with the Banque de France to put
in place the monetary policy tools for supporting price stability and
28 CAMEL-type regulatory tools combine ratios on capital adequacy, Asset quality,
Management, Earnings, and Liquidity to develop a composite rating score that is
used to assess the soundness of financial institutions.
29 According to the Tunisian authorities, an Early Warning System that permits the
BCT to rapidly intervene in banks in difficulty is being created.

32 | OVERVIEW OF THE BANKING SECTORS
reinforcing institutional capacities, including transparency of monetary
policy actions.
As the events in the beginning of 2011 amply demonstrate, political
stability is a key challenge for the country and the region as a whole. It is
therefore questionable to what extent these ambitious aims, especially those
relating to the branding of Tunisia’s banking sector as a regional centre, can
materialise without stable and sustainable political conditions. In addition,
the Tunisian authorities have to aim more at devising a forward-looking
supervisory and regulatory regime, giving banks the proper incentives and
ability to manage risks and limiting moral hazard.
1.6 Summary
The foregoing analysis reveals several common features of the banking
sectors of the Southern Mediterranean countries. In recent years, the
authorities of the four surveyed countries have engaged in a variety of
reforms to modernise their banking systems. These include restructuring
and privatisation of public banks, implementation of prudential regulation
and risk management frameworks and enhancing supervisory
responsibilities. Morocco and Egypt have improved the availability and
sharing of credit information. These reforms have led to a persistent
growth of credit to the private sector.
The discussion above shows that one potential explanation of
financial under-development is the heavy presence of the state, either
directly in the form of publicly-owned banks or indirectly in the form of
public debt in banks’ portfolios. For the latest years for which data are
available (2008-09), the market shares of public banks range from a low of
one-quarter of total banking assets in Morocco and Tunisia to highs of 67%
over 90% Egypt and Algeria, respectively. These ownership structures and
the underlying conditions, such as the high returns that government debt
earns in Egypt, are likely to crowd out the credit to private enterprises.
Indeed, public debt and loans, including loans to public enterprises,
account for nearly one-third of the total balance sheets of the Algerian and
Egyptian banks, surpassing the share of private credit.
Aside from crowding out private credit and constraining financial
development, the state’s dominant role in the banking sector appears to
have a serious negative impact on credit quality. Indeed, the ratios of nonperforming
loans to gross loans for the Southern Mediterranean countries
are among the highest globally. Owing to the relatively limited role of the
state, Morocco is once again an exception, with the lowest NPL ratios

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 33
among the four countries. Moreover, the four countries have implemented
policies to improve the quality of loans, including privatisation
improvements in credit information systems, loan repurchase programmes
and other plans to clean balance sheets. Nevertheless, the banking systems
of the four Southern Mediterranean countries have the worst loan qualities
in the region (Figure 1.4).
Figure 1.4 Non-performing bank loans in the Mediterranean and Middle East
(% of total loans, 2008)
Note: The 2008 figures on NPLs do not reflect substantial worsening in loan qualities in most
countries, notably Greece.
Sources: IMF, Global Financial Stability Report, October 2010 and Bank of Algeria.
The persistence of the non-performing assets and underdeveloped
financial systems remain leads to questions on the adequacy of the recent
regulatory reforms in the banking sector in the four countries covered in
this study. As noted above, the prevalence of the publicly-owned banks
may be at the root of the problem. However, shortcomings in various legal,
regulatory and supervisory frameworks may also matter. The next section
provides a deeper analysis of the regulatory conditions over several
dimensions, providing the analytical tools for making cross-country
comparisons over time.

2. CONVERGENCE OF BANKING SECTOR
REGULATIONS
T
he previous chapter has shown that all the four countries have faced
substantial reforms in their financial sectors in recent years. In this
chapter, a number of indices are developed in order to assess and
track the evolution of the adequacy of banking regulations using
publicly available and comparable surveys on banking regulations for a
large sample of countries since the early 2000s. To allow comparability
across the Mediterranean, the section develops the measures for a total of
11 Mediterranean countries, including five South-MED countries (Algeria,
Egypt, Israel, Morocco and Tunisia); and six EU-MED countries (Cyprus,
Spain, Greece, Italy, Malta and Portugal).
The aim of this section is to develop quantitative measures of
regulatory development that could serve as an indicator in the empirical
exercises that follow. Seven distinct regulatory areas are identified for
assessing different dimensions of regulatory adequacy. These cover
definition of banking, licensing requirements, capital requirements,
independence and power of supervisor, presence of safety nets, disclosure
and availability of credit information using distinct data sources. Although
these provide a broad view of the extent of regulation, several potential
areas (i.e. payment and settlement systems, credit guarantee schemes,
financial inclusion, etc.) have been excluded due to the unavailability of
comparable information sources for the sampled countries.
2.1 Methodology
The main source of information for the regulatory adequacy indices are the
Bank Regulation and Supervision Surveys (henceforth the ‘BRSS’)
| 34

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 35
developed by Barth et al. (2001), later revised in 2003 and 2007. 30 All three
surveys are built on official responses to questionnaires that were sent to
the national regulatory and supervisory agencies of over 120 countries,
most of which were returned. 31 The questions cover a wide variety of areas,
including banking activity, entry, capital regulations, supervisory
authority, private monitoring, deposit insurance and external governance.
One of the key advantages of the BRSS is that the questionnaires have
remained relatively similar over the years, although the later versions cover
more areas than the original survey. This particular feature of the datasets
allows us to make comparisons by building composite indices based on
specific answers over time to track the evolution of the different regulatory
and supervisory elements.
Figure 2.1 Average response rates for Bank Regulation and Supervision Surveys
(BRSS) of Barth et al.
Note: Response rates are averaged over the three surveys and correspond to the number of
questions with complete (i.e. excluding empty or partial) answers divided by the total
number of questions for that year.
Source: BRSS.
30 For the discussion of the results and other aspects of the data, see Barth et al.
(2006 and 2008).
31 The number of countries responding to the survey has increased over time. The
original survey of Barth et al. (2001) had 117 country respondents, including a wide
diversity of developed, developing and underdeveloped countries. The later
revisions achieved greater participation, with 152 in 2003 and 142 country
participants in 2007.

36 | CONVERGENCE OF BANKING SECTOR REGULATIONS
A key disadvantage of the Barth et al. (2001) survey is that the
number of questions responded to in the 2003 and 2007 revisions vary from
one country to another. For the Mediterranean countries, the aggregate
response rates are in generally greater than for the entire sample. As noted
in Figure 2.1, Morocco’s regulatory authorities have been the most
responsive to the survey, with nearly a 95% average response rate. In turn,
the other three Southern Mediterranean countries – Algeria, Egypt and
Tunisia – have achieved the lower response rates, at about 85%, which is
below the mean for the Mediterranean countries and comparable with the
average rate for the entire sample.
Although the response rates appear high in general, the existence of
even a single partial or empty answer renders the construction of a relevant
composite index dubious since there is no clear way of scoring for missing
responses. 32 Moreover, some countries, such as Tunisia and Algeria, have
not responded to all the three surveys, with Tunisia responding only in
2003 and Algeria in 2003 and 2007. To avoid any inconsistencies, all of the
indices used in this study are constructed using questions for which there
are complete (i.e. non-missing) responses. Moreover, the assessment of
regulatory convergence is based on the calculation of regional averages,
weighted by the total banking assets of each country. These allow us to
make a sounder judgment of whether the regulatory conditions on both
coasts of the Mediterranean are converging.
A second disadvantage of Barth et al. (2001) and its revisions was that
the questions did not cover all the regulatory and supervisory areas. Two
major areas where the surveys lacked depth were the details on deposit
insurance guarantee schemes and institutional variables, such as the extent
of credit information sharing and creditors’ legal rights. In order to cope
with this shortcoming, several additional sources were used to supplement
the construction of the composite indices, including the deposit insurance
database of Demirgüç-Kunt, Karacaovalı and Laeven (2005), IMF and
World Bank’s Financial Sector Assessment reports, World Bank’s Doing
Business Indicators and the websites of the national authorities.
In addition to the data from international organisations and national
authorities, a questionnaire of the Southern Mediterranean regulatory
agencies was also conducted to obtain deeper and more recent information.
32 Our approach differs from Barth et al. (2006), where empty answers were scored
as zero in the construction of the relevant indices.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 37
The questionnaires were of a quantitative and qualitative nature,
comprising an aggregate data collection exercise (section I), which was
completed by the country experts using official data and a face-to-face
interview (section II) with a senior official from the regulatory or
supervisory agency. In addition to completing some of the missing
elements of the Barth et al. (2001) surveys, the surveys also focused on
existing cooperation with the EU authorities as well as self-assessments on
foreign entry and competition, quality of audits, issues relating to the
application of Basel II standards, credibility of the insurance schemes and
other challenges. 33
The face-to-face interviews and data collection were carried out by
country experts Mohammed Yazid Boumghar (Algeria), Jawad Kerdoudi
(Morocco) and Moez Labidi (Tunisia). No interviews or data requests were
conducted in Egypt. For Tunisia and Morocco, the responses for the face-toface
interviews were low, with a large number of the requested items
remaining unanswered. A face-to-face interview could not be held with the
Algerian authorities, although a detailed summary of the financial market
developments and regulations were provided by the country expert (see
Table 2.1).
Table 2.1 Response rates to the CEPS questionnaire (% of questions responded)
Section I – Data request
Algeria Tunisia Morocco
1. Supervisory agency 70% 90% 0%
2. Entry & licensing 43% 49% 0%
3.Information disclosure 50% 100% 0%
4. Prudential requirements 22% 67% 0%
5. Crisis management 33% 100% 0%
6. Deposit insurance scheme 71% n.a. 0%
7. Market infrastructure 63% 100% 0%
TOTAL – Section I 49% 65% 0%
Section II - Interviews
33 The full set of questions contained in the questionnaire is reproduced in Annex 1.

38 | CONVERGENCE OF BANKING SECTOR REGULATIONS
1. Supervisory agency 0% 86% 64%
2. Entry & licensing 0% 83% 67%
3.Information disclosure 0% 75% 75%
4. Prudential requirements 0% 100% 79%
5. Crisis management 0% 95% 90%
6. Deposit insurance scheme 0% n.a. 88%
7. Market infrastructure 0% 75% 13%
8. Final remarks 0% 100% 75%
Total – Section II 0% 90% 72%
Total – Sections I & II 31% 75% 27%
Notes: See Annex for an entire list of questions covered under each area. There were no
responses to Sections I and II from the Moroccan and Algerian authorities, respectively. The
questionnaires were not sent to the Egyptian authorities. Since a deposit scheme is not in
place in Tunisia, the questions under part 6 were not applicable.
Seven composite indices are created using the various data sources
identified above, covering:
I. Scope restrictions
II. Entry obstacles
III. Capital requirement stringency
IV. Supervisory authority
V. Deposit insurance
VI. Private monitoring
VII. Credit information and laws
These areas provide a relatively broad coverage of the quality and
evolution of banking regulation and supervision. The composite indices
were calculated for each country and also for the South-MED (plus Israel)
and Euro-MED countries included in our sample.
The following sections revise and compare the evolution of the
regulatory conditions in each of the seven areas noted above.
2.2 Area I: Scope Restrictions
As is evident from the differing business models of financial institutions
across the world, financial institutions are growing increasingly complex
and offering a wider spectrum of products. Some countries restrict banking
to a narrow range of activities, such as taking deposits and issuing credit
with little flexibility in debt and asset management, while others provide

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 39
more flexibility. The regulations typically restrict the extent to which banks
may engage in the business of i) securities underwriting, brokering,
dealing, and all aspects of mutual fund industry; ii) insurance underwriting
and selling; and iii) real estate investment, development and management.
The composite indicator used in this area to assess the extent of
restrictions imposed on banking activity is based on the Banking Activity
Restrictiveness Index in BRSs. 34 The surveys provide measures on the
degrees of restrictiveness for each one of the above three categories,
ranging from unrestricted (1 point) and mostly permitted (2 points), to fully
prohibited (4 points). The Banking Activity Restrictiveness Index sums up
the scores for each category to come up a measure of the extent to which
restrictions are present on banks, with a maximum restrictiveness score of
12 points, where no activity other than narrow banking is allowed.
Table 2.2 Banking activity restrictiveness (% of maximum score)
2000 2003 2007
Algeria 42 50
Egypt 58 58
Israel 83 83 75
Morocco 83 58 75
Tunisia 67
SOUTH-MED* 83 72 68
Cyprus 42 67 67
Spain 50 42 42
Greece 58 67 50
Italy 58 67 75
Malta 58 67 67
Portugal 50 58 75
EU-MED* 55 57 59
Note: Greater values represent more restrictive rules as percent of maximum of 12 points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
The country-specific results summarised in Table 2.2 show that the
regulators in the South-MED impose more restrictions than the EU-MED
34 Banking activity restrictiveness index is constructed by summing up the scores
for the World Bank Guide (WBG) questions 4.1-4.3, as detailed in Appendix 2 of
Barth et al. (2006).

40 | CONVERGENCE OF BANKING SECTOR REGULATIONS
countries in general. A deeper analysis of the survey results (not included
here) shows that on both coasts of the Mediterranean, regulators impose
some form of restriction on insurance activities, although at a declining
extent over the years. Israel’s banks face the least flexibility among the
sampled countries, where all real-estate activities and some securities and
insurance activities are prohibited. Morocco’s banks cannot engage in realestate
investment, except for a brief period around 2003. In turn, Algeria’s
banks face few restrictions, with complete freedom to engage in securities
and real-estate investment activities. Egypt imposes some restrictions on
insurance and real-estate, largely comparable with the EU-MED countries.
Although it is not possible to judge the changing conditions in Tunisia due
to the lack of data for 2000 and 2007, the country remained close to the
South-MED averages. Turning to the EU-MED, while the banks in Spain
and Greece face fewer restrictions than their neighbours, there appears to
be an opposite tendency for the other countries, especially for Cyprus, Italy
and Portugal.
The figures show that there is a convergence tendency when the
regional weighted-averages are considered. Indeed, while the EU-MED
weighted-averages move up gradually over time, the South-MED averages
go down, converging on the former. However, there are clear differences
within each sub-region. For example, Israel imposes substantial restrictions
while Egypt has the most flexible system. As for the EU-MED, Spain’s
system imposes the least amount of restrictions while Italy has increasingly
narrowed the scope of banking activities over the years.
2.3 Area II: Entry obstacles
The competitive conditions in a country depend crucially on the regulatory
structures, and conditions hinder or prevent entry into the banking sector
by domestic or foreign banks. In some countries, the obstacles may take the
form of excessive licensing or entry requirements, which is applicable for
domestic and foreign banks together. In others, the governments may
restrict foreign entry as part of a conscientious policy choice, either
explicitly through setting limits on ownership or more importantly by
rejecting foreign applications in a disproportionate manner. 35 Lastly, a
35 Denials of domestic banks are not considered here as they are more likely to arise
from prudential concerns, including funding deficiencies or other financial
problems, which are common place for home-grown banks in countries with less
developed financial systems that have limited access to external capital.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 41
banking sector that is predominantly state-owned may be disadvantageous
for the development of privately-owned banks. 36
Three indicators are utilised to construct the composite index
assessing the impact of entry obstacles.
The first indicator that comes to mind for measuring how much the
regulatory structure obstructs entry are the legal licensing requirements,
which may hamper entry by making the procedures unnecessarily
cumbersome. The relevant measure is based on the set of requirements for
the licensing application to be considered valid. The index is built on the
total number of required documents, including: i) draft by-laws, ii)
organisational chart, iii) financial projections, iv) financial information on
potential shareholders, v) background of directors, vi) background of
management, vii) details of funding sources and viii) market differentiation
intended. 37
Table 2.3 shows that most South-MED countries impose similar levels
of stringency in terms of entry requirements with the EU-MED countries. In
particular, all of the eight requirements named above are commonplace in
almost all of the South-MED. In this respect, the entry requirements in
Israel can be clearly distinguished, where potential entrants are only
expected to submit financial projections and backgrounds of directors as
well as managers. As for the EU-MED, most countries require all or almost
all of the eight documents. Cyprus has also reduced its requirements in
recent years, where potential entrants are required only to submit draft bylaws
and the background of managers as well as directors.
Table 2.3 Entry into banking requirements (% of maximum score)
2000 2003 2007
36 Aside from their potentially negative impact on entry, state-owned banks may
fulfil an important developmental role in under-developed regions. Recent
evidence shows that in the Middle East and North Africa (MENA) region, public
banks compensate for the low private bank involvement in the SME sector,
engaging in more risky loan issuance, although they seem to have less than
sufficient capacity to manage such risks (Rocha et al., 2010b). See also Andrianova
et al. (2010) for recent evidence that government ownership of banks is associated
with higher long-run growth rates in developing countries.
37 The entry into banking requirements index is constructed by summing up the
scores for the World Bank Guide (WBG) questions 1.8.1-1.8.8, as detailed in
Appendix 2 of Barth et al. (2006).

42 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Algeria 88 100
Egypt 75 100 100
Israel 75 38 38
Morocco 100 100 100
Tunisia 100
SOUTH-MED* 78 71 71
Cyprus 100 75 38
Spain 100 88 88
Greece 100 100 100
Italy 100 100 100
Malta 88 88 88
Portugal 100 100 88
EU-MED* 99 98 92
Note: Greater values represent greater restrictive rules as share of a maximum of 8 points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
These results show that most countries in the Mediterranean require
similar documents for licensing. This means that these figures probably
gives at best an incomplete picture of the obstacles faced by potential
entrants. More realistically, these requirements are most likely used on
both sides of the Mediterranean to screen potential entrants, ensuring that
they are ‘fit and proper’ to run a banking business. In contrast, in Israel and
Cyprus, the bar is set much lower, possibly to attract foreign and domestic
banks to set up their offices on these countries to facilitate entry.
As noted above, the set of licensing requirements do not paint a
complete picture of entry obstacles. The second index considers the more
discretionary power that the authorities enjoy by granting or rejecting
entry. More specifically, the index is based on the fraction of foreign
banking licensing applications that have been denied within the past five
years from the day the questionnaire was conducted. 38 The relevant data
are only available for the 2003 and 2007 questionnaires.
Table 2.4 clearly shows that foreign banking application denials are
more commonplace in the South-MED countries, which is in stark contrast
with the EU-MED where such denials are very rare. 39 In particular, all of
38 Share of foreign denials are addressed by World Bank Guide (WBG) question
1.10, as detailed in Appendix 1 and 2 of Barth et al. (2006).
39 The responses to our own survey do not reveal any particular reasoning—
political or otherwise—for the elevated foreign denial rates in the four Southern-
Mediterranean countries. In their response to the relevant question on foreign
entry (Question 2.3.a, see Annex), the Tunisian authorities have stated that

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 43
the (four) foreign banking licensing applications for the years between 1998
and 2002 have been denied in Egypt. More recently, Egypt has denied
nearly a third of the foreign licensing applications (13 out of 41) within the
five years leading to 2007. Morocco has denied one of the two applications
at the same period. Israel has also refused nearly one-fifth of all the foreign
applications, potentially offsetting its relatively relaxed licensing
requirements as noted above. Algeria and Tunisia do not appear to use
foreign denials as an entry obstacle. These results show that foreign denials
could be one place where there is little sign of convergence between some
of the South-MED and EU-MED countries.
The third and last indicator on entry obstacles relates to the
dominance of government-controlled banking. The index is a simple
measure of the market power of banks that are majority-owned by the
state, i.e. 50% or more, in terms of total assets. 40
Table 2.5 points at significant differences on both sides of the
Mediterranean. While the state has little control over banking in the EU-
MED countries, except for Portugal and Greece, public banks represent
between 30 to 90% of the banking activity in the South-MED. This is
particularly the case for Algeria and Egypt where the state has a control
over a significant majority of the banking sector. State-owned banks in
these countries often enjoy implicit or explicit state guarantees, with access
to public funding and possibly subject to less strict or flexible rules, which
may be a disadvantage for potential entrants and more generally
undermining competition (Barth et al., 2004).
although foreign entry could be advantageous on bank governance and
accumulation of ‘know-how’, it can also serve to elevate risks by facilitating the
transmission of external shocks to the domestic financial system.
40 Share of government-controlled banks is addressed by World Bank Guide (WBG)
question 3.8.1, as detailed in Appendix 1 of Barth et al. (2006).

44 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.4 Share of foreign applications denied
2003 2007
Algeria 0%
Egypt 100% 32%
Israel 17% 20%
Morocco 50%
Tunisia 0% 0%**
SOUTH-MED* 39% 27%
Cyprus 0% 0%
Spain 7% 0%
Greece 14% 0%
Italy 13% 3%
Malta 0% 0%
Portugal 0% 0%
EU-MED* 10% 1%
Notes: Question not included in the 2000 questionnaire.
* Regional averages are weighted by total banking assets.
** For Tunisia, the 2007 result obtained from own survey.
Source: BRSS.
Table 2.5 Market share of government-controlled banks (% of total assets)
2003 2007
Algeria 96 90
Egypt 65 67
Israel 46 0
Morocco 35 29
Tunisia 43
SOUTH-MED* 55 33
Cyprus 4 3
Spain 0 0
Greece 23
Italy 10 9
Malta 0 0
Portugal 23 25
EU-MED* 8 6
Notes: Figures represent share of banks with at least 50% state ownership.
* Regional averages are weighted by total banking assets.
Source: BRSS.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 45
Put together, the three indices provide a contrasting picture of the
sampled countries in terms of entry obstacles. The set of documents needed
for a valid licensing application are similar on both sides of the
Mediterranean to large extent. These requirements are most likely used to
ensure that only ‘fit and proper’ undertakings are allowed to operate as
banks. Only two countries, Cyprus and Israel, can be distinguished in this
respect, with few licensing requirements. Turning to less official controls
that the authorities exert on the banking sector, foreign entry denials are
proportionally high in some of the South-MED countries, particularly in
Egypt, Israel and Morocco. The state also maintains a substantial direct
control over the banking sector in most of the countries in the region, with
publicly owned banks accounting for more than two-thirds of the banking
sector activities in Algeria and Egypt. In short, although the official entry
conditions appear comparable, there are significant and persistent entry
obstacles that can curtail competition in the South-MED banking sectors,
possibly emanating from official authority in practice and political
interference.
2.4 Area III: Capital requirement stringency
One of the common aims of regulating banks is to ensure that they operate
soundly. Regulatory capital requirements are an important part of these
rules, which determine the minimum amount of capital a bank should hold
relative to its total assets.
Comparing the capital ratios represents a good first step towards
understanding how sound the banking sector is. There are clear signs that
the capital ratios have converged over time. For example, as depicted in
Table 2.6, the total capital ratios have converged over time across both sides
of the Mediterranean over time. First of all, with the exception of Malta,
which can be considered as an offshore centre, all the countries have
maintained a total capital ratio of between 9 to 15%. 41 In recent years, the
banks in South-MED countries have actually become better capitalised,
with the average capital ratios reaching to 13% in 2009.
41 The banks in the so-called ‘offshore financial centres’ often provide a number of
advantages to their clients, including low taxes, light regulation and account
anonymity. These side benefits enable them to collect deposits at relatively low
costs and place them in safe and liquid assets. As a consequence, these banks often
have above-average capital adequacy ratios; see Barth et al. (2006, pp. 173-177,
especially Figure 3.27), for evidence.

46 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.6 Regulatory capital ratios (% of risk weighted assets)
1998 2001 2005 2009
Algeria 11.9 12.0
Egypt 10.2 9.8 14.1 15.1
Israel 9.2 9.5 10.7 12.6
Morocco 13.1 12.6 11.5 11.8
Tunisia 11.1 12.4 12.4
SOUTH-MED* 10.0 10.3 12.1 13.1
Cyprus 9.9 14.0
Spain 12.5 13.0 11.7 12.2
Greece 11.4 13.6 13.3 11.7
Italy 13.4 10.4 10.6 12.1
Malta 15.3 18.4 20.6 23.9
Portugal 12.3 9.5 11.3 10.5
EU-MED* 12.9 11.0 11.3 12.0
Notes: Figures represent share of total capital in risk-weighted assets using the 1988 Basle
Accord definitions.
* Regional averages are weighted by total banking assets.
Sources: BRSS and IMF Global Financial Stability Reports (GFSR).
The Southern-MED banks appear to be at least as well-capitalised as
their Northern counterparts, especially after early 2000s. Does this result
reflect the stringency of capital requirements or a lower appetite for risk? In
other words, is it the regulations that make the banks sounder or are the
banks simply not willing to take too many risks? In order to answer this
important question, it is necessary to look deeper into the rules.
There are different ways of measuring the stringency of capital
requirements. The index that is used here gives consideration to the types
of capital allowed, the risk-weights applied, and whether the minimum
capital ratios vary with risk. More specifically, the capital stringency index
aims to determine the extent to which capital requirements restrict leverage
potential and risky behaviour, including questions on i) whether the
minimum capital-to-asset requirements are in line with 1988 Basle Accord
definitions; ii) whether the minimum ratio varies with the bank’s credit risk
or iii) market risk; and whether the value of iv) unrealised loan losses, v)
unrealised security losses or vi) foreign exchange losses are deducted from
regulatory capital. Additionally, the index aims to measure the restrictions
imposed on the source of regulatory capital, such as vii) whether these
funds are verified by regulatory authorities; and, whether viii) cash and
government securities, or more generally ix) non-borrowed funds are the

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 47
only allowed forms of capital for initial disbursements and subsequent
injections. 42 A greater number of affirmative responses to these questions
lead to a higher stringency score.
Table 2.7 summarises the comparison of the stringency of the capital
requirements for the countries in our sample. A quick glance through the
figures reveals a contrasting picture. Among the South-MED countries,
Algeria is a clear exception according to the results of the 2007 survey, with
affirmative answers to all questions except the variability of minimum
capital ratio according to an individual bank’s market risk. The capital
requirements are relatively flexible for other countries in the South-MED.
For example, in Egypt minimum capital ratios do not vary and the
unrealised loan, security or foreign exchange losses are not deducted from
regulatory capital. The same also is the case for Morocco and, to a lesser
extent, Israel. 43 The 2003 rules in Tunisia appear to be comparable, at least
according the 2003 survey for which data are available, to the EU norms.
Among the EU-MED countries, Spain has the most stringent capital
requirements, with affirmative answers to all of the nine questions in 2003
and 2007, followed by Portugal, Malta and Cyprus. For the latter two, there
is a clear tendency of substantial strengthening of rules following their EU
accession in 2004. Italy has the most lenient capital requirements, where the
minimum capital ratios are constant for all banks, only unrealised securities
losses are deducted from regulatory capital and there are no restrictions on
the source of regulatory capital as noted in the questionnaire.
42 The stringency of capital requirements index is addressed by World Bank Guide
(WBG) questions 3.1.1, 3.2, 3.3, 3.9.1, 3.9.2, 3.9.3, and 1.5—1.7. The calculation of the
index is detailed in Appendix 2 of Barth et al. (2006), pp. 337-338. One question
(WBG 3.7) on the fraction of revaluation gains allowed as part of capital is omitted
from the calculation of the index since the responses were not available for most
countries in our sample.
43 Morocco has recently enacted laws to make its capital requirements more
stringent. In 2009, the minimum capital ratio was raised from 8% to 10%. The
Moroccan authorities are also preparing to apply an even-higher capital adequacy
requirement of 12% to the more risky undertakings.

48 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.7 Stringency of capital requirements (% of maximum score)
2000 2003 2007
Algeria 33 89
Egypt 56 33 33
Israel 33 67 44
Morocco 44 44 33
Tunisia 67
SOUTH-MED* 43 51 45
Cyprus 11 44 67
Spain 89 100 100
Greece 44 56 33
Italy 33 22 22
Malta 67 56 67
Portugal 44 78 89
EU-MED* 54 57 62
Note: Greater values represent greater restrictive rules as a share of a maximum score of 9
points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
With these results in hand, it is easy to see that there is a pattern of
divergence. Some of the EU members have exceptionally flexible capital
requirements, while the opposite is true for others, especially the new
member states. In turn, capital requirements of most of the South-MED
countries are in general less stringent than the EU-MED averages,
especially regarding rules on deductions for unrealised loan losses and
risk-based supervision. 44 Therefore, the capital adequacy ratios are high in
the South-MED, most probably not because of the stringency of the
underlying rules but because of the business models and the risk-aversion
of banks.
The findings should be interpreted with care. The capital requirement
standards as summarised by the Basle Accords were designed, at least until
44 These results are largely in line with the key regulatory shortcomings identified
for the region in Tahari et al. (2007), using compliance of European countries with
Basel Core Principles on prudential regulations and requirements (BCPs 6 to 15) as
a benchmark.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 49
very recently, for and by the regulators in advanced economies. Our own
surveys in Morocco and Tunisia reveal that the regulators face constant
challenges in applying the capital requirements. The key difficulties
identified by the authorities are instructive and include: i) absence of
external rating systems; ii) deficiencies in banks’ information systems,
which favour foreign banks over domestic institutions; iii) lack of welldeveloped
credit information systems; iv) financial opaqueness in most
enterprises and v) resistance to change, both from banks and borrowers.
There is also evidence that the strict enforcement of capital adequacy
requirements can lead to a severe ‘credit crunch’ in countries where the
alternatives to bank-based financing are less developed (Chiuri et al., 2002).
In turn, all of the Southern Mediterranean countries, with the exception of
Morocco, have substantial stocks of non-performing loans, which warrant a
more severe approach to provisioning practices.
These remarks highlight the fact that the regulations that are
conceived with developed countries in mind may not always be applicable
for developing countries, calling for refinements in certain areas,
including—but not limited to—the applicability of the internal-ratings
based approach to risk in Basel II.
2.5 Area IV: Supervisory authority
A key issue in the effectiveness of banking regulations is whether the
supervisory authorities have the necessary powers to apply a variety of
measures to discipline or, at the extreme, resolve banks that violate the
rules or engage in imprudent activities. To that extent, in most countries,
the supervisors take prompt corrective action against a bank if the capital
falls below the minimally required level. If the deterioration of the bank
continues, the supervisor must have the ability to find a resolution before
the bank becomes insolvent, posing a systemic threat. In order to be
effective, the supervisors need access to reliable and frequently updated
information on the condition of the banks. The judicial systems often allow
the courts to intervene, diminishing, postponing or reversing illegitimate
supervisory actions; however these should not undermine the supervisor’s
chief responsibility of protecting and ensuring an orderly operation of the
banking market. These aspects of the supervisory system issues should be
in line with the regulatory priorities and not subject to political patronage.
In short, the supervisors should have the authority to discipline potentially
troubled banks and resolve problems while remaining independent from
political influence.

50 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Two indices are used for measuring supervisory authority.
The first index measures the official power of the supervisor to take
specific actions to correct or prevent problems. The relevant questions
include the ability of supervisors to i) meet external auditors without
approval of bank; ii) communicate directly with auditors on illicit activities
undertaken by bank’s management or directors; iii) receive disclosure of
off-balance sheet items; iv) take legal action against negligent auditors; v)
change organisational structure of troubled banks; vi) order management
or directors to cover losses; vii) suspend dividend distributions; viii)
bonuses and ix) management fees. Additionally, for the 2003 and 2007
surveys, additional questions on troubled banks were also considered on
the supervisors’ ability to x) declare insolvency; xi) suspend ownership
rights; xii) supersede shareholder rights and xiii) fire or hire management
or xiv) directors. 45 An affirmative answer to any of these questions
represents a greater supervisory power. Some of these powers may only be
exercisable by some supervisory-like institutions, such as the depository
insurance agency or the bank restructuring agencies, which grant a more
moderate power to supervisors. 46 In other cases, the courts or the
government may be involved, which would serve to void the power of the
supervisors in those actions.
Interestingly, Table 2.8 shows that the South-MED grant more power
to their supervisory authorities. This is particularly the case in Egypt,
which has responded affirmatively to all of the questions in all three
surveys. In Morocco, the supervisory authorities also exert substantial
power, although they cannot take legal action against external auditors for
negligence. In Algeria, which is the only South-MED country with
decreasing official power in 2007, the supervisory agency may no longer be
able to suspend the granting of executive bonuses and management fees for
troubled banks. In Tunisia, only the courts can declare a bank insolvent,
which is increasingly the case in most developed countries.
45 The official supervisory power index is addressed by World Bank Guide (WBG)
questions 5.5-5.7, 6.1, 10.4, 11.2, 11.3.1-11.3.3, 11.6, 11.7, and 11.9.1-11.9.3. The
calculation of the index is detailed in Appendix 2 of Barth et al. (2006), pp. 339-342.
46 In these cases, the aggregate score is augment by only ½ points; for more details,
see calculation of the index is detailed in Appendix 2 of Barth et al. (2006), pp. 339-
342.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 51
Table 2.8 Official supervisory power (% of maximum score)
2000 2003 2007
Algeria 100 79
Egypt 100 100 100
Israel 44 50 71
Morocco 78 89 93
Tunisia 93
SOUTH-MED* 69 76 83
Cyprus 100 57 86
Spain 44 64 82
Greece 56 86 71
Italy 33 36 50
Malta 67 100 100
Portugal 67 100 100
EU-MED* 41 54 69
Note: Greater values represent greater restrictive
rules as share of a maximum score of 14 points in
2003 and 2007 and 9 points in 2000.
* Regional averages are weighted by total banking assets.
Source: BRSS.
Turning to the EU-MED countries, it is interesting to see that the new
member states, Cyprus and Malta, grant increasing official power to their
authorities. The same applies to Portugal and, to a lesser extent, Spain. Italy
once again obtains the lowest score in official supervisory power: unlike
other countries in our sample, the Italian supervisory authority has no right
to sue external auditors for negligence, order directors or managers to
cover losses, suspend the decision to distribute dividends, bonuses, or
management fees, or remove the management.
The second index for assessing supervisory authority turns more
generally to the independence of the supervisor from political influence.
For this index, three questions from the BRSS are considered: i) Are
supervisory bodies accountable only to a legislative body? ii) Are
supervisors legally liable for its actions committed in exercise of their
duties? iii) Does the head of the agency have a fixed term? The level of
independence is determined by points obtained by counting affirmative
answers to questions (i) and (iii) and a negative answer to (ii). 47
47 The independence from political interference index is addressed by World Bank
Guide (WBG) questions 12.2, 12.10, and 12.2.2. The calculation of the index is
slightly different than the specification in Appendix 2 of Barth et al. (2006), pp. 349-

52 | CONVERGENCE OF BANKING SECTOR REGULATIONS
The results depicted in Table 2.9 show a clear divergence in terms of
independence from political interference. While the banking supervisors of
the EU-MED countries have become more independent, not much has
changed in the South-MED countries. The biggest concern remains the
accountability of the supervisor directly to the executive arm, i.e. president,
prime minister or other cabinet members, which is the case in all of the
Southern Mediterranean countries. 48
Table 2.9 Independence from political interference (% of maximum score)
2003 2007
Algeria 33 0
Egypt 67 67
Israel 33 33
Morocco 33 33
Tunisia 67
SOUTH-MED* 44 36
Cyprus 67 100
Spain 67 100
Greece 67 67
Italy 0 33
Malta 100 67
Portugal 100 100
EU-MED* 37 70
Notes: Greater values represent more independence as share of a maximum score of 3 points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
Of particular concern is Algeria, where none of the three criteria
outlined above is satisfied in 2007, which implies an enormous potential for
political interference. The same can also be said for other countries, such as
Israel, Morocco and possibly Tunisia. In comparison, the supervisor is
accountable to the Parliament in almost all EU members except Italy and
Malta. Once again, the Italian supervisory authority remains well below the
350, in that in order to score a point in question 12.2 the supervisory bodies should
be accountable to no one other than a legislative body, such as the Parliament or
the Congress.
48 In the case of Morocco, the governor of the Bank Al-Maghrib serves at the
discretion of the King.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 53
EU standards in terms of independence from political interference due to
its accountability to the central government and its legal liability for
damages to a bank in exercise of its duties. Another key distinguishing
factor is the fixed-term for the head of the regulatory authority, which is
not available as an option in Algeria, Israel or Morocco but has become
increasingly popular among the EU members.
The results of the BRSS surveys reviewed in this section show that the
powers granted have increased or remained constant in almost all of the
countries. Moreover, the official powers granted to supervisors appear to
be on the rise on both sides of the Mediterranean. Turning to operational
independence, the government officials have the ability to politically
interfere in the work of the supervisors. Therefore, despite the fact that the
supervisors are assigned almost full authority, it is possible that these
powers remain notional due to government interference. Provided that
some of the South-MED countries have substantial government presence in
the banking sector (already noted above), the operational independence
should be a guiding principle to ensure that all banks – publicly or
privately-owned – are treated equally.
2.6 Area V: Deposit insurance
Deposit insurance systems are among the key elements of a country’s
financial safety net, designed to prevent any disruptions to the financial
markets and the economy. By protecting depositors, the deposit insurance
schemes provide confidence to relatively small depositors and prevent
bank runs. At the same time, they may introduce moral hazard,
diminishing the depositors’ incentives to monitor and screen the banks and
amplifying the shareholders’ incentives to engage in excessive risk. The
moral hazard problem implies that banks have incentives to take on risk
that can be shifted to a deposit insurance fund or, ultimately, the tax
payers.
Efforts are being taken across the world to mitigate moral hazard
problems arising from deposit guarantee schemes. 49 First, the amount of
coverage matters. In some countries, aside from limits on the total amount,
a co-insurance is imposed to ensure that depositors bear some part of the
49 See Kane (2000) and Demirgüç-Kunt et al. (2005) for a review of the potential
effects and key design features of the deposit insurance schemes.

54 | CONVERGENCE OF BANKING SECTOR REGULATIONS
costs. 50 Second, the use of risk-adjusted premiums may also serve to better
internalise the costs of the risks that they take. Third, the way that the
deposit insurance schemes are funded also matters. For example, when the
government is explicitly or implicitly involved in providing the necessary
funds, moral hazard may be attenuated, especially in countries where the
government has ample resources. In turn, when the system is backed with
funds by banks, moral hazard can be limited by the understanding that the
amount of guarantees is restricted with the pooled reserves.
Looking at the existing schemes, there are clear differences on both coasts
of the Mediterranean (Table 2.10). The revised EU Deposit Insurance
Directive requires member states to maintain deposit insurance with a
coverage limit of at least €100,000, raised from a minimum of €20,000 in the
aftermath of the 2007-09 financial crisis. 51 Most of the countries in the EU-
MED have chosen to set this base amount as their coverage limits,
representing between to 4 to 7 times the average annual income figures.
The 2009 amendment has also abolished the co-insurance system, which
allowed up to 10% of losses to be shared with covered depositors. Riskbased
premiums exist only in Italy and Portugal. Setting itself clearly apart
from the other countries in the region, Italy has an ex-post funding
structure, where the banks are required to contribute after the deposit
guarantee scheme is activated. Cyprus and Malta have hybrid systems in
which substantial amounts of supplementary (ex-post) funding may be
activated if the funds’ resources fall below pre-set levels. The levels of exante
funds display substantial variation, wherever they exist, with a low of
0.01% of eligible deposits in Cyprus and a high of 1.00% in Portugal.
50 Empirical evidence shows that the coverage limits and co-insurance practices
serve to reduce bank failure likelihoods substantially, (Demirgüç-Kunt &
Detragiache, 2002).
51 Directive 2009/14/EC, which amended the Deposit Guarantee Directive
94/19/EC. The minimum amount of €100,000 has been in force as of 31 December
2010.

55 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.10 Deposit guarantee schemes in the Mediterranean, latest available figures
Est. Coverage limit
Primary Co-
Risk-based Ex-post/ Ex-ante
date € (Dec.
2010)
(% of GDP per
capita, PPP)
funding
by
insurance premiums ex-ante coverage
ratio**
SOUTH-MED
Algeria 1997 6,200 108% Banks No No Ex-ante n. .
Egypt -
Israel -
Morocco
Tu isia
EU-MED
2 06 7,200 228% Banks No N Ex-ante 1.40%
Cyprus 1997 100,00 4 3% Banks No* No Hybrid 0.01%
Spain 1977 100,000 439% Banks No* No Ex-ante 0.80%
Greece 1995 100,000 482% Banks No* No Ex-ante 0.58%
Italy 1987 103,291 462% Banks No* Yes Ex-post 0.00%
Malta 2003 100,000 785% Banks No* No Hybrid 0.10%
Portugal 1992 100,000 595% Banks No* Yes Ex-ante 1.00%
* Co-insurance has been abandoned by the amending Directive 2009/14/EC.
** The actual coverage ratio is calculated as the ratio of ex-ante funds and eligible deposits using published figures for 2007-08.
Sources: European Commission (2010), IMF (2008), Bank Al-Maghrib (BAM) and Banque d'Algérie (BNA) .

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 56
Tuning to the South-MED countries, Egypt, Israel, and Tunisia have
no schemes in place. 52 In Algeria and Morocco, the coverage limits
represent one to two times the average annual incomes, pointing at a much
lower level of protection afforded than in the EU. As in the EU-MED, the
deposit guarantee schemes do not have a co-insurance option or use riskbase
premiums. Algerian authorities are involved in the funding of the
system, which is not surprising since the publicly-owned banks represent
nearly 90% of the total banking assets.
The deposit insurance scheme index identifies the level of observance
of standards that are thought to mitigate the moral hazard problem. Since
recent information is available, the index is constructed for the years 2003,
2007 and 2010. For countries with an explicit system, three issues are
relevant: i) whether a co-insurance discount is applicable to payouts, ii)
whether premiums are risk-adjusted and iii) whether only banks take a
primary role. 53 An additional point is scored for an affirmative answer to
each one of these questions. A score of zero is assigned to countries where
no explicit system exists, since in those cases the government is assumed to
provide implicit guarantees, implying a greater incentive to take risks by
banks. 54
52 In Egypt, although the legal framework allows for the establishment of an
autonomous deposit insurance fund, no scheme has been setup yet.
53 The calculation of the deposit insurance scheme index follows the format
detailed in Barth et al. (2006, p. 354), except that a score of zero is assigned for
countries with no explicit insurance scheme.
Three separate sources were used for the deposit insurance scheme information.
First, the BRSs provided the basic information and evaluation for 2003 and 2007.
Whenever the BRSS gave conflicting or incomplete results, the information
contained in Demirgüç-Kunt et al. (2005), the European Commission’s (2010)
assessment of EU deposit guarantee schemes as well as the legal documents from
the websites of Bank Al-Maghrib (BAM) and Banque d’Algérie (BNA) were used.
54 Gropp & Vesala (2004) shows that credible implicit guarantees operating
through the expectation of public intervention at times of distress can aggravate
the moral hazard problem when compared to explicit deposit guarantee schemes.
As the authors note, the key issue is whether the institutional and fiscal conditions
would make the inherent guarantees credible. It is assumed here that the three
countries with no explicit systems, namely Egypt, Israel and Tunisia, have ample
fiscal resources and the necessary institutional framework that could make such
guarantees credible.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 57
Table 2.11 Deposit insurance index (% of maximum score)
2003 2007 2010
Algeria 0** 33 33
Egypt 0 0 0
Israel 0 0 0
Morocco 0 33 33
Tunisia 0 0 0
SOUTH-MED* 0 8 8
Cyprus 67 33 33
Spain 33 33 33
Greece 33 33 33
Italy 67 67 67
Malta 67 33 33
Portugal 100 100 67
EU-MED* 55 52 50
Notes: Greater values represent greater restrictive rules as share of a maximum score of
3points.
* Regional averages are weighted by total banking assets.
** The Algerian deposit guarantee system, which existed since 1997, was partly funded by
the government in 2003.
Sources: BRSS, Demirgüç-Kunt et al. (2005), European Commission (2010), Bank Al-Maghrib
(BAM) and Banque d’Algérie (BNA).
The figures in Table 2.11 show that moral hazard issues are more of a
threat in the South-MED countries. For the most part, this is due to the
absence of deposit guarantee schemes in Egypt, Israel and Morocco (in
2003), and Tunisia. The Algerian system was equivalent to an implicit
guarantee in 2003 since the government had a direct funding role. 55
Turning to countries with explicit systems, some similarities emerge. Out of
the three issues outlined above, Algeria, Cyprus, Spain, Greece, Malta and
Morocco only satisfy the requirement that the banks (and not the
government) take the primary role of funding the scheme in 2010. The
Italian and Portuguese systems, in turn, include risk-adjusted premiums,
impacting significantly the EU-MED averages. Lastly, the EU-MED
55 Under Law no. 90-10 of 1990 regarding money and credit, the Algerian Treasury
was a contributor to the deposit guarantee fund (Art. 170). More recently, the
government’s funding role has been replaced with full funding by banks under the
amending Law no. 03-11 of 2003 regarding money and credit (Art. 118).

58 | CONVERGENCE OF BANKING SECTOR REGULATIONS
averages display a downward trend, which is entirely due to the gradual
abandonment of the co-insurance payouts.
Many of the South-MED countries have been reluctant to develop
deposit insurance schemes. A badly designed scheme can invite additional
risks and may not be better than a system with no scheme at all. The results
show that the schemes that exist in Algeria and Morocco (as well as in other
EU-MED countries) may indeed amplify the moral hazard risks. These
conclusions, however, should be interpreted with care. As the recent
financial crisis has shown, when a run on a bank has the potential to spur
broader panic, the governments are likely to step in to stop a potential bank
run, notwithstanding the type of explicit arrangements in place. 56 One may
wonder, quite justifiably, whether the named arrangements do really
mitigate moral hazard when they may be so easily replaced with limitless
state support. However, it should not be forgotten that such blanket
guarantees are not viable in most of the South-MED countries with limited
public resources. Therefore, the explicit schemes, wherever they exist, are
the only viable insurance for depositors, highlighting the importance of the
design issues in resource-poor countries.
2.7 Area VI: Private monitoring
Most of the regulatory factors considered in this study relate to the rules
and standards set forth by the regulators, which are used to distinguish
between acceptable and unsound behaviour. In this manner, the regulatory
principles are often well-defined, calling for compliance with specific rules
or standards. However, banks are also influenced by these hard-wired
forces. Market forces and investors may also be crucial in shaping the
decisions and, in particular, restraining risky behaviour. For example,
block-holders can, at least in theory, exercise their voting power to
influence managerial actions. More realistically, debtors or stockholders
use available information to assess the bank’s conditions and indirectly
influence the management by withdrawing funds, which has an impact on
the borrowing costs of the banks. As far as depositors and other debtholders
are concerned, private monitoring could be seriously undermined
when an explicit and overly generous deposit insurance scheme exists.
56 This was amply demonstrated during the Northern Rock fall of 2007 when the
UK Treasury, extended the existing guarantees on bank deposits – with a
maximum payout of £31,700 at the time—to cover all deposits.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 59
The availability of reliable and timely information to investors is at
the core of market disciple. The index is therefore based on the survey
responses to a number of questions on disclosure rules and standards,
comprising whether: i) a certified audit is required; all of top-10 banks are
rated by ii) domestic and iii) international credit rating agencies; income
standards include accrued through unpaid interest on iv) performing or v)
non-performing loans; vi) banks are required to produce consolidated
accounts; vii) directors are liable for erroneous or misleading reporting;
viii) subordinated debt allowable or required as part of capital; ix) offbalance
items are disclosed to the public; x) banks are required to disclose
risk management procedures and xi) supervisors are required to make
enforcement actions public. 57 The private monitoring score increases with
affirmative answers to the previous set of questions.
The comparisons points at a small but growing disparity between the
coasts of the Mediterranean (Table 2.12). Although most countries fulfil a
majority of the requirements, the constant progress of the European
countries is not paralleled in Southern countries.
The most striking difference between the Southern and Northern
countries is the share of the top-10 banks that are rated by (international or
domestic) credit rating agencies, which has widened substantially
according to the 2007 survey. In particular, almost all of the top-10 banks
are rated by credit rating agencies in the EU, except Malta. In the South-
MED countries, most banks are not rated. In some cases, this is due to the
inherent structure of the market. For example, Algeria’s largest banks are
state-owned and were not subject to ratings as of 2007. In other countries,
there are clear problems with disclosure. In two of the most developed
markets in the region, Israel and Morocco, only half of the top-10 banks are
rated. 58
57 The private monitoring index is addressed by World Bank Guide (WBG)
questions 5.1, 5.3, 10.7.1-2, 10.1, 10.1.1, 10.3, 10.6, 3.5-6, 10.4.1, 10.5, and 11.1.1. The
calculation of the index is slightly different than the specification in Appendix 2 of
Barth et al. (2006, pp. 350-352), excluding a question on the presence of an explicit
deposit insurance, which is already covered in another index.
58 These results may also arise from a small or highly concentrated banking sector.
In such a case, only a handful of top banks will dominate the banking sector while
the other (smaller) banks will be subject to less investor scrutiny.

60 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.12 Private monitoring (% of maximum score)
2000 2003 2007
Algeria 73 64
Egypt 64 73 73
Israel 73 82 82
Morocco 64 82 73
Tunisia 55
SOUTH-MED* 68 77 71
Cyprus 73 82 82
Spain 82 82 91
Greece 64 82 91
Italy 64 82 82
Malta 73 82 82
Portugal 82 64 73
EU-MED* 71 80 86
Note: Greater values represent greater
restrictive rules as share of a maximum score of 11 points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
Another common issue, especially more recently, is the exclusion
accrued (though unpaid) interest from income statements, which allows
them undue flexibilities in determining their earnings. Also, Tunisian and
Algerian banks are not required to produce consolidated accounts that
cover all financial subsidiaries. Lastly, according to the 2003 BRSS, the
banks in Tunisia are not required to make public their risk management
procedures, which became standard in the region in recent years.
These results show that the regulatory structures of South-MED
countries have not matched the progress in the North countries in
enhancing their disclosure rules. It is true that there are broad similarities
on both sides of the Mediterranean. For example, a certified audit is
compulsory in all of the sample countries and the accounting rules exhibit
similarities in most of the countries. However, the proportion of banks
subject to independent ratings has not changed much in the South-MED
countries over the past few years. These results call for a serious
examination of the readiness or the appropriateness of deposit insurance
coverage in these countries.
2.8 Area VII: Credit information and laws
Access to information and creditor protection laws are crucial for ensuring
the smooth operation of credit markets. Economic theory suggests two

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 61
crucial limits to the amount of credit that financial institutions can grant to
potential borrowers. On the one hand, credit conditions are clearly bound
by the ability of creditors to enforce contracts, require repayment, claim
collateral and possibly gain control over the receivables. The easier these
actions, the more likely will be the lenders to grant the loans. On the other
hand, lenders would like to have access to accurate information on the
potential borrowers, such as credit histories, other lenders and other
banking transactions.
Theoretical models suggest that an operational information-sharing
infrastructure can reduce adverse selection in credit markets and facilitate
access to credit, especially among more opaque borrowers such as small-
and medium-size enterprises (SMEs) (Pagano & Jappelli, 1993). When such
information is available, the creditors can make a better judgement on the
credit-worthiness of the borrowers. Other studies have documented the
importance of creditors’ rights on the availability of credit (La Porta et al.,
1998 and Levine, 1998). Recent studies have confirmed these views with
increasingly convincing evidence that both credit information mechanisms
and creditors’ rights have a nontrivial impact on the flow of credit and
financial development (Jappelli & Pagano, 2002; Djankov et al., 2007; and
Haselmann et al., 2010).
The credit information and laws indices developed in this subsection
are based on the Getting Credit methodology developed in the World
Bank’s Doing Business surveys. 59 The relevant area covers the legal rights
of borrowers and lenders with respect to secured transactions and the
extent of credit information-sharing. Two sets of indicators are used for
these purposes.
The first set describes how well the collateral and bankruptcy laws
facilitate lending, covering: i) ability to use moveable assets while keeping
possession of assets; ability to obtain non-possessory security rights in ii) a
single or iii) all moveable asset classes without requiring a specific
description of the collateral; iv) extension of security rights to future or
59 First started in 2003, the World Bank’s Doing Business surveys cover over 180
countries, providing a snapshot of regulatory and legal conditions and their effects
on businesses, especially on small and medium-size enterprises (SMEs). Each year,
the surveys are sent out to a large number of local experts specialising in different
fields, including lawyers, consultants, officials and other professionals who are in
close contact with the legal and regulatory structures of the covered countries (the
results of the surveys are available from http://www.doingbusiness.org/).

62 | CONVERGENCE OF BANKING SECTOR REGULATIONS
after-acquired assets; v) ability to secure all types of debts and obligations
via a general description; vi) availability of a collateral registry; ability of
secured creditors to obtain priority without exceptions in the case of vii)
defaults viii) liquidations, and ix) restructuring; and x) possibility of out-ofcourt
agreements on collateral enforcement. An affirmative answer to any
one of these questions enhances the relevant scores. 60
Table 2.13 Strength of legal rights (% of maximum score)
04-05 07-08
Algeria 30 30
Egypt 30 30
Israel 90 90
Morocco 30 30
Tunisia 30 30
SOUTH-MED* 57 55
Cyprus 90
Spain 60 60
Greece 30 30
Italy 30 30
Malta
Portugal 30 30
EU-MED* 41 44
Note: Greater values represent more
independence as share of a maximum score of 10 points.
* Regional averages are weighted by total banking assets.
Source: BRSS.
Table 2.13 shows that the legal rights granted to creditors are slightly
less in the South. Israel does exceptionally well, better than almost all
countries, by satisfying all but one criterion on the availability of out-ofcourt
agreements on collateral enforcement. Among the EU-MED
countries, Cyprus also does equivalently well, complying with all but one
criterion, namely regarding the secured creditors’ claims during
reorganisation. Spain also performs well but fails to grant some of the
sought-after rights to secured creditors and over future assets. Other
countries, including those in the South-MED, do relatively badly,
complying only with the standards on the use of movable assets as
60 See the World Bank’s Doing Business website on more details on the
methodology (http://www.doingbusiness.org/methodology/getting-credit).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 63
collateral, ability to grant non-possessory rights for a group of assets and
the use of debts in collateral agreements.
The second index measures the availability, coverage and depth of
credit information, either through public credit registries or private credit
bureaus. The relevant questions relate to the i) collection both positive and
negative information, ii) collection of data on firms and information, iii)
collection of data from retailers and utility companies, iv) availability of
credit history for at least two years, v) availability of data on small loans
(i.e. less than 1% of annual incomes) and vi) ability of borrowers to access
their credit history. As above, an affirmative answer to any one of these
questions leads to an additional score for the credit information index.
Table 2.14 clearly shows that the South-MED countries lag behind
their Northern counterparts in terms of the depth of credit information. The
figures also show that the differences have diminished in recent years.
Israel is clearly an outlier, especially according to the more recent Doing
Business survey where it satisfies all of the six criteria except the
distribution of positive and negative credit information. With no credit
information sharing infrastructure in place, Cyprus is another exception.
More broadly, the EU-MED countries comply with almost all of the
criteria. A common shortcoming, present in Italy, Greece and Portugal, is
that the private registries do not collect information from retailers or utility
companies. The South-MED countries (except Israel) have notable
deficiencies on credit information availability and sharing, despite
significant improvements over the last few years. As of 2008, two countries
in the region had operational private credit bureaus: Egypt and Israel.
According to the results of the 2007-08 Doing Business survey, Egypt scores
relatively low since borrowers have no right to inspect their credit histories
and since data from retailers are not used. 61
61 Egypt has substantially improved on these weaknesses in 2009-10 and has
increased its score to a perfect 100%, according to the Doing Business 2011 survey
results. For more information, see http://www.doingbusiness.org/reforms/
overview/economy/egypt.

64 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.14 Depth of credit information (% of maximum score)
04-05 07-08
Algeria 17 33
Egypt 33 67
Israel 50 83
Morocco 17 17**
Tunisia 33 50
SOUTH-MED* 37 62
Cyprus 0
Spain 83 83
Greece 67 67
Italy 100 83
Malta
Portugal 83 83
EU-MED* 91 82
Note: Greater values represent more
independence as share of a maximum score of 6 points.
* Regional averages are weighted by total banking assets.
** The 2007-08 figures for Morocco do not take account of the fact of the creation of the new
private credit bureau that became operational in 2009.
Source: BRSS
Public credit registers exist in all of the countries, although their
effectiveness varies. Focusing on the South-MED countries with public
registries only, i.e. Algeria, Morocco and Tunisia, several common
weaknesses are notable. 62 First, none of the countries provide borrowers
the right to access or contest information on their own credit histories.
Second, the registries do not collect and distribute data from retail and
utility companies. Also, Tunisia and Morocco also scores relatively low,
mostly due to the absence of detailed data collection. Lastly, although the
Algerian public registry collects and distributes both positive and negative
information, it does not distribute detailed information on a borrower’s
history or small loans.
In summing up, the figures above show that despite substantial
reforms in recent years, the South-MED countries clearly lag behind in
terms of the use of credit information. The same cannot be said concerning
62 According to the most recent Doing Business 2011 survey for the year 2010,
Morocco obtains an almost perfect score (83%). For more information, see
http://www.doingbusiness.org/reforms/overview/economy/morocco.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 65
the strength of legal rights; most of them on both sides of the
Mediterranean have a similar set of legal rights granted to the creditors.
Based on recent findings that highlight the importance of credit
information-sharing not only for the availability of credit but also for the
stability of the banking sectors as a whole, the South-MED countries should
do all they can to converge to the EU’s standards on this front (Houston et
al., 2010). The types of reforms that have been introduced in several
countries, such as Egypt, should be a model for others in the region to
assure an even development of their economies with the flow of credit
flowing to the smaller and more opaque firms.
2.9 Conclusions
The previous sections reviewed the quality and the level of convergence of
the regulatory and supervisory structures of the South-MED and EU-MED.
The assessment included seven dimensions, including scope of banking,
entry obstacles, the stringency of capital requirements, the power and
independence of the supervisory authority, incentives provided by the
deposit insurance scheme, private monitoring and creditors’ rights and
access to information. This section will provide a summary of these areas,
offering a comparative analysis of the seven composite indicators that
aggregate the relevant indices.
Figure 2.2 and Table 2.15 diagramme and summarise the key
remaining weaknesses that distinguish the South-MED countries from their
counterparts in the North.

66 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Figure 2.2 Regulatory standards in South-MED and EU-MED regions
Note: The diagrammes above sum up the weighted averages for the regulatory indices in
each of the seven areas discussed in sections 2.2 to 2.8. The North-South disparities are
highlighted in shades of red, with darker shades representing greater differences, i.e. more
than 25% disparity. For entry into banking, depth of credit information and strength of legal
rights indices, the South-MED averages exclude Israel as an outlier. For credit information
and laws indices, 2004-05 and 2007-08 figures were used.

67 | CONVERGENCE OF BANKING SECTOR REGULATIONS
Table 2.15 Key regulatory weaknesses in the South-MED
AREA I. Scope
restrictions
AREA II. Entry
obstacles
AREA III.
Capital
requirements
AREA IV.
Supervisory
authority
AREA V.
Deposit
insurance
AREA VI.
Private
monitoring
AREA VII.
Credit info. &
laws
Source: Authors’ compilation.
Description General remarks Algeria Egypt Morocco Tunisia
Restrictions on or
prohibition of
various activities
Licensing, foreign
entry & presence of
public banks
Extent to which
capital requirements
restrict risks
Ability of
supervisors to
prevent & correct
problems
Presence of an
explicit scheme &
mitigation of moral
hazard
Availability of
reliable & timely
information to
investors
Ability of legal &
information systems
to facilitate lending
In line with EU-
MED standards
Below EU-MED
standards due to
foreign denials &
role of government
Increasing disparity
due to riskinsensitivity
Below EU-MED
standards due
potential for
political interference
Below EU-MED
standards due to
implicit insurance &
adverse incentives
Increasing disparity
due to poor
accounting practices
Below EU-MED
standards due to
deficient info system
Some restrictions on
insurance
Public banks
represent >90% of
bank activity
Surpassing most of
EU-MED standards
High potential for
political interference
No co-insurance or
risk-adjusted
premiums
Top banks not rated;
flexibility in
accounting
Public registry only;
no borrower access
or detailed info
Some restrictions on
real-estate
Foreign denials;
public banks
represent > 60% of
bank activity
Market & credit
risks not considered;
broad def. of capital
Supervisor enjoys
full set of powers
Implicit government
guarantees
Several top banks
not rated
Private registry
established in 2006
Some restrictions on
insurance; realestate
activities
prohibited
Some restrictions on
insurance & realestate
activities
Foreign denials Few obstacles;
public banks have
diminishing role
Market & credit risk
not considered;
broad def. of capital
Some potential for
interference
No co-insurance or
risk-adjusted
premiums
Several top banks
not rated; flexibility
in accountings
Public registry only
prior 2009; common
issues prior that
date
Comparable with
EU-MED standards
High potential for
political interference
Implicit government
guarantees
Flexibility in
accounting rules; no
risk mgt. disclosure
Public registry only;
no borrower access
or detailed info

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 68
The collective assessment of the convergence of the regulatory and
supervisory structures of the South-MED countries with the EU-MED
standards gives a mixed picture (Figure 2.2 and Table 2.15). Despite some
improvements, key weaknesses remain in deposit insurance, entry
obstacles and credit information. Moreover, some recent issues recent
disparities have also become apparent, especially in the stringency of
capital requirements, potential for political interference and private
monitoring.
The deposit insurance index has failed to improve since neither the
Egyptian nor Tunisian authorities have put in place an explicit insurance
scheme. As discussed in Section 2.6, implicit schemes may enhance risktaking
through a blanket government guarantee for the leading institutions.
Moreover, even in Algeria and Morocco, no effort has been made to align
the banks’ incentives by implementing risk-based premiums or coinsurance
schemes, which would help internalise some of the costs to the
deposit guarantee schemes due to excessive risk-taking.
The South-MED countries have implemented a number of reforms to
improve the availability and use of credit information by financial
institutions. Egypt and, more recently, Morocco have established private
credit bureaus in 2006 and 2009, respectively. Nevertheless, the South-
North gap has not been narrowed. Algeria and Tunisia continue to rely
only on public registries, restrict the borrowers’ right to inspect their credit
histories, fail to collect and distribute detailed data, including from nonbank
sources, such as retail stores or utility companies. Although the
literature provides little guidance, private credit bureaus have an improved
access to new technologies and know-how to ensure that informationsharing
mechanisms work effectively. The countries in the region should
continue to monitor developments and spearhead innovative systems to
use the stock of information and infrastructure already set-up by the public
systems. 63
Another major issue, the presence of entry obstacles, continues to be a
key weakness of the regulatory structures of the region. Although the
licensing requirements exhibit similarities on both sides of the
63 Morocco may serve as interesting example, by effectively combining the data
collection roles and capacities of the Bank Al-Maghrib, which operates the public
registry, and the newly established private credit bureau, Experian-Morocco. For a
comparative analysis of the Moroccan and Egyptian credit information systems,
see Madeddu (2010, pp. 21-23).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 69
Mediterranean, other indicators point at substantial barriers to entry.
Government ownership, which is widespread in the region, gives undue
advantages to incumbent banks and restricts entry incentives. In Morocco,
the government-owned banks represent a declining proportion of total
bank activities; in Algeria, Egypt and (to some extent) in Tunisia, the
government-ownership persists. Although government ownership may
have some beneficial side benefits, the authorities have to ensure that such
roles are well-defined and should not be an obstacle to the development of
the financial system. 64 The rates of foreign denials are also very high,
further supporting the idea of substantial entry barriers and competitive
advantages enjoyed by domestic incumbent banks.
In addition to these three key weaknesses, the 2007 survey points at
three new concerns. The stringencies of capital requirements, which were
in line with the EU standards in 2003, have deteriorated according to the
latest BRSS survey. There are some exceptions, like Algeria and, to a lesser
extent, Tunisia. However, in Egypt and Morocco, the capital requirements
and accounting standards have become more flexible and less risksensitive.
Poor accounting practices have also contributed to an increasing
disparity in private monitoring indices.
Lastly, political interference has become a significant possibility,
potentially undermining supervisory authority and reinforcing the
governments’ direct control – an additional concern on the competitiveness
and efficiency of the banking sector. As the eruption of public discontent in
Tunisia and Egypt in early 2011 clearly attests, the region’s governments
have attempted to maintain (perhaps for far too long) a tight grip on their
countries’ political and economic systems. It is exactly such forms of
interference that may conflict with the objectives of the financial and
competition authorities.
64 Rocha et al. (2010a) notes the essential role that public banks fulfil in the region
by providing financing to the SMEs. The authors note that private banks are unable
to fill this gap largely due to the generally weak quality of financial infrastructure,
including the availability and reliability of information on potential borrowers.

3. ANALYSIS OF EFFICIENCY AND
CONVERGENCE
T
he process of financial reform undertaken by both developed and
developing countries aimed to establish a market-based financial
sector, to boost bank competition through improved mobilisation of
savings, to enhance market-based allocation of resources and to
foster more efficient risk-management capabilities. However, the
conventional wisdom relating to the positive effect of reforms on financial
sector performance is not always validated by empirical studies (Berger et
al., 2000). Despite a vast literature on the effects of deregulation on the
efficiency and productivity of banks (see Berger & Mester, 2003; Mukherjee
et al., 2001, Isik & Hasan, 2003, Zhao et al., 2010, among others)
deregulation seems to have had a positive effect in some countries but not
in others. Indeed, the outcome of deregulation policies seems to reflect
several country-specific demand and supply conditions of the banking
industry prior to deregulation.
This part of the study attempts to shed light on these issues by
examining the effect of financial reform on the efficiency of the banking
sector in 11 countries in the Mediterranean region: Cyprus (CY), Algeria
(DZ), Egypt (EG), Spain (ES), Greece (GR), Israel (IL), Italy (IT), Malta (MT),
Morocco (MA), Portugal (PT) and Tunisia (TN) over the period 1995-2008.
The second part of the analysis aims to contribute to the current debate on
fostering integration in the Mediterranean region. Following Casu &
Girardone (2010), we use the concepts of β-convergence and σ-convergence
and employ a dynamic panel data analysis to assess the speed at which
financial markets are integrating.
Our results indicate an improvement in bank efficiency across the
region, particularly in the latter part of the sample period. The overall mean
efficiency in the region is increasing, driven by technological improvements
by the best practice banks. Spanish banks dominate the region both in
| 70

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 71
terms of overall efficiency and of meta-technology ratios. Nonetheless,
during the sample period, the average meta-technology ratio for the region
is also increasing, thus indicating an ability of banks in all countries to
appropriate the best available technology.
These results are supported by the estimation of β-convergence σconvergence.
The β coefficient is always negative and statistically
significant, thus indicating that convergence in efficiency scores has
occurred across countries in the MED-11 area. Furthermore, results for the
σ-convergence suggest an increase in the speed of convergence as the σ
coefficient is always negative and statistically significant. This indicates
that, whereas the technological gap is still wide, it is narrowing at a faster
speed.
3.1 Literature review
There is a vast literature on the use of frontier techniques to evaluate bank
efficiency, using both parametric and non-parametric methodologies.
While earlier studies focused on one methodological approach and on
individual countries (mainly the US, or EU countries) (Berger &
Humphrey, 1997; Goddard et al., 2001), in recent years both the number of
cross-country studies and the number of studies focusing on developing
countries has increased, mainly due to the unprecedented economic
reforms implemented in such countries (for a review of recent literature
see, among others, Berger (2007), Goddard et al. (2007) and Hughes &
Mester (2010).
Most cross-country studies assume that banks in different countries
can access the same banking production technology. In other words, they
assume a common production frontier for all countries in order to be able
to compare efficiency results across borders. The interpretation of the
resulting efficiency scores relies significantly on the validity of this
assumption. In some cases this is a major drawback, as the production
technology is substantially different among countries, particularly if
countries are at different levels of financial development. Bank efficiency
estimates may be influenced by factors not generally included in the
efficiency analysis, such as differences in bank type, ownership and other
bank specific conditions. In such cases, the assumption of a common
frontier may be misleading. Further, such an assumption can lead to bias
efficiency results of banks from different countries as it ignores differences
in regulatory, competitive and economic conditions that are beyond a
bank’s control (Dietsch & Lozano-Vivas, 2000 and Chaffai et al., 2001). The

72 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
Bos & Kool (2006) study indicates that if environmental factors are not
appropriately controlled, efficiency estimates may be biased. Recent
empirical studies have attempted to overcome this problem by integrating
country-specific environmental variables into the efficiency estimation.
The influence of environmental variables on cross-country efficiency
levels has been of interest for many researchers. Bikker (2004) investigates
the differences in X-efficiency levels of European banks and concludes that
X-efficiency estimates from single-country studies, as often found in the
literature, can be very misleading. He documents significant differences in
cost-efficiency scores across countries and sizes of banks, bank
specialisation as well as institutional conditions (supervisory rules,
government interference, customer preferences and level of development).
Bos et al. (2005) analyse the effects of accounting for heterogeneity on the
German bank efficiency scores for the period 1993-2003. They find that
banks of different sizes, geographic origins and types (cooperative and
savings) have significantly different cost efficiency scores. Dietsch &
Lozano-Vivas (2000) investigated the influence of the environmental
conditions on the cost-efficiency of the French and Spanish banking sector
over the period 1992-98. They showed that the specific environmental
conditions of each country occupy an important role in the definition and
specification of the common frontier of different countries.
In fact, when environmental variables are incorporated in the model,
the differences between both banking industries are reduced substantially
and the cost-efficiency scores improved. The Chaffai et al. (2001) study, on
a sample of European countries over the period 1993-97, report similar
findings. They conclude that controlling for environmental conditions
reduces the differences in average operational inefficiency scores among
countries. Grigorian & Manole (2006) use a DEA approach to estimate the
efficiency levels of transition countries between 1993 and 1995 and a twostep
approach to explain the differences in efficiency across countries. They
find that foreign ownership and enterprise restructuring enhance
commercial bank efficiency. Bos & Kool (2006), on the other hand, find that
market specific factors and regional macroeconomic factors are of limited
importance in explaining operational efficiency of the Dutch cooperative
banking sector. Battese et al. (2004) have recently proposed a so-called
‘meta-frontier’ as the method to estimate country or regional-specific
frontiers and end up with efficiency scores that can be compared in an
absolute sense. The meta-frontier results from the envelopment of regional
specific frontiers. Bos & Schmiedel (2003) apply the meta-frontier
methodology to eight European banking markets for the period 1993–2000.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 73
The authors conclude that for most countries included in the study, profit
efficiency in particular improves significantly when estimated using a
meta-frontier instead of a common frontier arguing that this may be
evidence of the importance of local market circumstances. Ben Naceur et al.
(2011) examine the effect of financial-sector reform on bank performance in
selected Middle Eastern and North African (MENA) countries in the period
1994-2008 using Data Envelopment Analysis (DEA) and employ a metafrontier
approach to calculate efficiency scores in a cross-country setting.
They then employ a second-stage regression to investigate the impact of
institutional, financial and bank-specific variables on bank efficiency.
Overall, their results show that the observed efficiency levels of banks vary
substantially across markets, with differences in technology explaining
most of the efficiency differentials.
Several studies investigate the existence and implications of financial
convergence, especially in relation to the deregulation processes. 65 Only a
few studies, however, directly address the issue of the relationship between
financial integration and bank efficiency. Tortosa-Ausina (2004) examines
the convergence in efficiency of Spanish banks following deregulation
through a model of distribution dynamics and finds evidence of decreased
dispersion of efficiency scores at the end of the deregulation period.
Murinde et al. (2004) investigate the convergence of the banking systems in
Europe following the launch of the single market programme in 1993. They
find weak evidence of convergence and only for specific products. Weill
(2009) attempts to provide evidence of financial integration by estimating
the convergence of cost efficiency derived from the application of SFA
methodology. His results indicate an on-going process of convergence at
the EU level. More recently, Casu & Girardone (2010) evaluate the
dynamics of EU banks’ cost efficiency by means of DEA and then assess
their convergence both towards an EU-wide frontier and towards best
practice. Their results seem to provide supporting evidence of convergence
of efficiency levels towards an EU average. Nevertheless, the potential
gains brought about by increased integration seem to have been offset by a
decrease in the overall efficiency levels of EU banks.
65 See Baele et al. (2004) for a review of different measures of financial market
integration.

74 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
3.2 Methodology
3.2.1 Data Envelopment Analysis
DEA is a mathematical linear programming technique developed by
Charnes et al. (1978), which identifies the efficient frontier from the linear
combination of those units/observations that (in a production space) use
comparatively fewer inputs to produce comparatively more outputs. The
original (or Charnes, Cooper and Rhoades - CCR model) assumes constant
returns to scale (CRS), which is the optimal scale in the long run. Banker,
Charnes and Cooper (1984) (or the BCC model) include an additional
convexity constraint (λ) to allow for variable returns to scale (VRS). In
particular, if at any time t there are N firms that use a vector of inputs to
produce a vector of outputs, the input-oriented BCC measure of efficiency
of a particular firm is calculated as:
where is the scalar efficiency score for the i-th unit. If =1 the
i-th firm is efficient as it lies on the frontier, whereas if < 1 the firm is
inefficient and needs a (1- ) reduction in the inputs levels to reach the
frontier.
The choice of using a DEA is based on several considerations: it
works well even with a small sample size and it does not require any
assumption about the functional form of the frontier or of the inefficiency
component. We adopt an input-minimisation orientation, based on the
assumption that during periods of regulatory changes and increased
competition, market participants strategically focus on cutting costs.
(1)

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 75
Therefore we would expect changes in inputs used to be closely associated
with the changes in market structure.
3.2.2 Meta-frontier analysis
There are various ways to incorporate environmental variables in the
estimation of bank efficiency, the most commonly used are the one-step
and the two-step approach. In the one-step approach, environmental
variables are included directly in the estimation of efficiency whereas in the
two-step approach, efficiency scores obtained in the first stage of analysis
are then regressed on a number of country-specific environmental
variables. Both approaches are employed in the literature: the one-step
approach seems to be the preferred choice if using a parametric approach
to the efficiency evaluation, following the maximum likelihood procedure
of Battese & Coelli (2005). On the other hand, the two-step approach seems
to be the favoured approach if efficiency is estimated by means of Data
Envelopment Analysis (DEA). In a typical two-stage study, the relative
efficiency of each institution is evaluated in the first stage based and then
regressed (as the dependent variable in an ordinary least squares or a Tobit
regression) on various explanatory variables in the second stage to identify
the factors whose impact on efficiency is statistically significant. A
theoretical justification for the use of a two-stage method that uses DEA in
the first stage is provided by Banker & Natarajan (2008).
Departing from the standard two stage approach, Battese et al. (2004)
and O’Donnell et al. (2008) recently proposed a so-called ‘meta-frontier’
method to estimate country or regional-specific frontiers and obtain
comparable efficiency scores, as the meta-frontier results from the
envelopment of regional specific frontiers.
In this study, to accommodate the potential country variation of
available banking technology and to obtain comparable technical
efficiencies for the countries in our sample, we follow the meta-frontier
approach. If we consider the available technology to be a state of
knowledge in existence at a given point in time, we can define the metatechnology
as the totality of the regional/country-specific technologies. The
meta-frontier production function is therefore a frontier function that
envelops all frontiers of individual countries/groups.
To apply the meta-frontier approach with DEA, it is necessary to
solve separate models (equation 1) for each country in order to specify the
country-specific frontiers and one for the joint data set for solving the metafrontier.
The efficiencies measured relative to the meta-frontier can be

76 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
decomposed into two components: a component that measures the distance
from an input-output point to the group frontier (the common measure of
technical efficiency) and a component that measures the distance between
the group frontier and the meta-frontier (representing the restrictive nature
of the production environment).
The meta-technology ratio (DEA-MTR), that is the relative
productivity of technologies, can be obtained as the ratio between metafrontier
(in)efficiency (DEA-M) and the country-specific (in)efficiency
(DEA-C). The higher the ratio, the closer a country’s production technology
is to the ‘best practice’ in the region. Vice versa, the lower the ratio, the
bigger is the technology gap.
3.2.3 Modelling convergence
To investigate the convergence of bank efficiency levels across the 11
countries in the Mediterranean region (MED-11) over the period 1994-2008,
we follow Casu & Girardone (2010) and employ the concepts of βconvergence
and σ-convergence (Barro & Sala-i-Martin, 1991, 1992 and
1995; and Quah, 1996).
To estimate unconditional β-convergence or the ‘catch-up effect’, we
employ the following equation:
where i=1,…11 and t=1,…14; = the mean efficiency of the banking
sector of country i at time t; = the mean efficiency of the banking
sector of country i at time t-1; ; α, β and ρ are the
parameters to be estimated and εi,t = error term. A negative value for the
parameter β implies convergence; the higher the coefficient in relative
terms the greater the tendency for convergence. Equation (2) is first
estimated without including the lagged dependent variable ( ), as in
the conventional growth theory models. The β-convergence equations are
estimated by pooled OLS regression and Generalised Method of Moments
(GMM) to introduce dynamic behaviour in the time series and crosssectional
variation (Blundell & Bond, 1998).
To estimate cross sectional dispersion or σ-convergence, that is to
estimate how quickly each country’s efficiency levels are converging to the
(2)

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 77
average, we adopt the following autoregressive distributed lag model
specification: 66
where ; ; and
are defined as before; the mean efficiency of the MED-11 banking
sectors at time t; the mean efficiency of the MED-11 banking sectors
at time t-1; ; α, σ and ρ are parameters to be calculated and
εi,t is the error term. σ < 0 represents the rate of convergence of towards
; the larger is σ in absolute value, the faster the rate of convergence. The
model in equation (3) is estimated initially without the inclusion of the
lagged dependent variable ( ), as we did for the β-convergence in
equation (2).
3.3 Descriptive statistics
The sample relates to a balanced sample of commercial and savings banks
in the following 11 countries in the Mediterranean region: Cyprus (CY),
Algeria (DZ), Egypt (EG), Spain (ES), Greece (GR), Israel (IL), Italy (IT),
Malta (MT), Morocco (MA), Portugal (PT) and Tunisia (TN) over the period
1995-2008.
Table 3.1 below provides some descriptive statistics of the total
number of observations in the sample and the average total assets by year
and country.
66 Similar specifications have been estimated, among others, by Fung (2006), Parikh
& Shibata (2004), Weill (2009) and Casu & Girardone (2010).
(3)

Table 3.1 Number of observations and average total assets (€ million)
N. of
obs.
CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 78
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Change
1995-2008
CY 4 2,369 2,895 3,371 3,653 4,505 5,435 6,585 7,053 7,652 8,653 10,489 14,221 18,209 21,300 899%
DZ 2 2,851 2,739 2,440 2,477 2,574 2,547 2,642 2,354 2,297 2,266 2,716 2,848 3,002 4,021 141%
EG 14 1,459 1,571 1,803 1,962 2,403 2,777 3,114 2,784 2,169 2,229 3,129 3,511 4,146 4,321 296%
ES 65 10,264 11,275 12,658 13,605 15,184 18,663 19,919 19,818 21,749 30,402 36,931 41,634 47,559 51,800 505%
GR 8 7,812 8,690 10,035 10,972 13,497 15,804 16,956 17,248 17,723 18,848 22,279 27,218 33,131 38,674 495%
IL 5 3,489 3,655 3,666 3,579 4,798 5,771 6,346 4,615 4,074 4,072 4,765 5,305 5,631 6,462 185%
IT 80 3,049 3,347 3,592 3,926 4,078 4,386 4,693 5,057 5,113 5,694 6,232 6,795 7,246 8,429 276%
MA 4 2,320 2,334 2,318 2,634 2,973 3,274 3,390 3,509 3,685 4,965 6,095 7,444 8,658 10,537 454%
MT 4 1,008 1,162 1,314 1,474 1,670 1,837 1,975 2,137 2,203 2,281 2,395 2,692 2,904 3,161 314%
PT 10 8,910 9,617 10,845 12,500 14,495 18,857 20,559 20,928 22,458 23,646 26,185 28,842 31,887 34,340 385%
TN 10 1,088 1,121 1,078 989 1,163 1,379 1,544 1,453 1,404 1,435 1,575 1,633 1,733 1,941 178%
Mean 206 4,056 4,401 4,829 5,252 6,122 7,339 7,975 7,905 8,230 9,499 11,163 12,922 14,919 16,817 375%

79 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
As is apparent from Table 3.1, Italy and Spain dominate the sample,
while the number of observations for Algeria, Israel, Morocco and Malta is
particularly low. Data availability improves in the final years of the sample
period, probably due to better reporting of accounting data; however, for
the purpose of the present analysis, we concentrate on continuously
operating institutions over the time period.
Substantial differences in the average size of banks are apparent, with
Spanish banks being the largest. The average size (total assets) of all
institutions in the sample increases from €4,056,000,000 in 1995 to
€16,817,000,000 in 2008. It is necessary to point out, however, that in some
countries the high number of small- and medium-sized institutions has an
impact on the overall country averages (for example, Italy), whereas in
other countries only a small number of large banks is present in the sample
(for example, Israel).
3.3.1 Input and output variables
There are two main approaches to the definition of inputs and outputs of
financial institutions: the production approach and the intermediation
approach. Both approaches are widely used in the literature and there is no
consensus on the superiority of one or the other. In this study we follow a
variation of the intermediation approach (Sealey & Lindley, 1977). This
approach views financial institutions as mediators between the supply and
the demand of funds. As a consequence, deposits are considered as inputs,
and interest on deposits as a component of total costs, together with labour
and capital.
In the cross-country setting of the present study, the need for
comparable data from different countries imposes strong restrictions on the
variables one is able to use, not least because of the various accounting
criteria used in the four countries under investigation. To minimise
possible bias arising from different accounting practices, the broad
definition of variables as presented by Bankscope was chosen.
Specifically, the input variable used in this study is Total Costs
(Interest Expenses + Overheads), whereas the output variables capture both
the traditional lending activity of banks (total loans) and the growing nonlending
activities (other earning assets).

80 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
We aggregate the cost expenditure 67 into a single input to minimise
the well-known dimensionality problem associated with DEA. In small
samples, if we have a high number of variables relative to the number of
observations, units can be wrongly identified as efficient because too many
constraints have been specified. Observations tend to become incomparable
and hence figure on the frontier owing to the inability of DEA to indentify
peers. One way around this, commonly used in the literature, is to
aggregate the input variables in a single monetary value. Tables 3.2, 3.3 and
3.4 report the descriptive statistics of the input variable (total costs) and the
two output variables (total loans and total other earning assets),
respectively.
Table 3.2 Total costs (average, € million)
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Change
1995-
2008
CY 179 221 254 288 320 392 418 464 465 494 547 615 913 1,130 630%
DZ 149 163 141 124 104 137 127 81 65 65 106 72 102 92 62%
EG 116 133 150 159 195 240 252 193 138 135 208 247 275 354 305%
ES 909 940 885 913 894 1,216 1,312 1,131 961 1,020 1,397 1,706 2,277 2,852 314%
GR 899 987 1,026 1,165 1,282 1,386 1,118 992 903 940 980 1,297 1,867 2,496 278%
IL 209 171 275 306 361 428 564 242 246 240 314 230 244 278 133%
IT 297 306 283 267 231 261 276 267 258 252 261 307 376 463 156%
MA 129 137 135 147 155 179 178 186 178 261 277 269 326 397 307%
MT 55 66 76 86 109 110 115 106 99 92 96 105 120 132 238%
PT 815 799 781 919 883 1,123 1,276 1,130 1,105 1,176 1,215 1,408 1,793 2,247 276%
TN 68 73 56 64 69 84 96 93 85 87 102 108 120 122 180%
Mean 348 363 369 404 418 505 521 444 409 433 500 579 765 960 276%
Total costs increase steadily (+276%) over the sample period; costs
increase in all countries. The largest increase is displayed by banks in
Cyprus (+630%) and the smallest by banks in Algeria (+ 62%).
Tables 3.3 and 3.4 show the averages for the two output variables,
total loans and total other earning assets.
67 Overheads comprise Personnel Expenses, Other Administrative Expenses and
Other Non-Interest Costs.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 81
Table 3.3 Total loans (average, € million)
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Change
1995-
2008
CY 1,334 1,639 1,903 2,185 2,610 3,155 3,592 4,019 4,360 4,841 5,484 7,474 10,327 13,413 1005%
DZ 863 792 602 429 539 690 842 734 763 784 939 893 994 1,358 157%
EG 656 740 865 971 1,270 1,506 1,604 1,309 902 884 1,160 1,233 1,430 1,658 253%
ES 4,619 5,183 6,169 7,021 8,072 9,967 11,027 11,622 13,112 18,454 22,804 28,097 32,289 34,605 749%
GR 2,734 3,110 3,516 4,224 5,437 6,679 7,926 9,427 10,569 11,477 13,410 17,193 22,395 28,460 1041%
IL 2,481 2,535 2,422 2,434 3,167 3,938 4,504 3,386 2,885 2,707 2,978 3,237 3,373 3,907 158%
IT 1,480 1,522 1,678 1,902 2,178 2,506 2,713 2,972 3,219 3,501 3,916 4,381 4,946 5,938 401%
MA 1,233 1,259 1,217 1,420 1,624 1,718 1,724 1,690 1,759 2,390 3,270 3,926 4,956 6,365 516%
MT 500 656 711 775 851 918 961 997 1,034 1,106 1,141 1,312 1,454 1,648 330%
PT 3,629 4,125 5,161 6,948 8,791 11,937 13,462 14,291 14,657 15,632 16,757 18,706 21,665 24,134 665%
TN 722 711 672 700 782 979 1,103 1,076 1,037 1,027 1,110 1,142 1,167 1,325 183%
Mean 1,841 2,025 2,265 2,637 3,211 3,999 4,496 4,684 4,936 5,709 6,634 7,963 9,545 11,165 606%
Total loans are steadily increasing in all countries, with the highest
percentage change displayed by Greek and Cypriot banks. This might be a
consequence of increased lending following EU membership. Other earning
assets are also increasing steadily, but not at the same rate of loans
(percentage change 230% versus 606%).
Table 3.4 Total other earning assets (average, € million)
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008
Change
1995-
2008
CY 685 878 1,094 1,038 1,238 1,545 2,221 2,072 2,416 2,963 4,117 5,220 6,242 6,009 877%
DZ 1,577 1,620 1,508 1,707 1,655 1,476 1,217 1,017 916 865 1,102 1,329 1,303 1,780 113%
EG 693 724 824 863 962 1,089 1,305 1,296 1,156 1,242 1,810 2,104 2,483 2,346 338%
ES 4,731 5,105 5,383 5,320 5,503 6,425 6,383 6,045 6,485 9,888 11,731 11,049 12,163 13,258 280%
GR 4,559 4,993 5,742 6,061 7,029 7,994 7,736 6,773 6,115 5,456 6,640 6,988 6,853 6,334 139%
IL 818 851 669 655 985 1,044 999 653 633 860 1,388 1,732 1,997 1,646 201%
IT 1,298 1,536 1,593 1,673 1,500 1,462 1,467 1,507 1,325 1,596 1,897 2,000 1,829 1,822 140%
MA 600 601 609 683 962 1,229 1,436 1,369 1,394 1,687 1,849 2,341 2,465 2,891 482%
MT 411 408 487 571 660 744 827 942 973 1,010 1,074 1,171 1,187 1,330 323%
PT 4,285 4,447 4,509 4,281 4,034 5,108 5,008 4,613 4,902 5,939 7,549 7,997 8,091 7,625 178%
TN 104 139 135 189 238 274 294 260 250 293 329 345 412 386 371%
Mean 1,797 1,937 2,050 2,095 2,251 2,581 2,627 2,413 2,415 2,891 3,590 3,843 4,093 4,130 230%

82 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
Figure 3.1 illustrates the trend of the input and output variables, both
as averages and as percentage change (1995 as base year). Bank growth in
these countries has been mainly driven by development of the traditional
lending function, and the increase in total costs seems to be mirrored by the
steady growth of bank loans. Cypriot and Moroccan banks display the
most remarkable increase in other earning assets; this is possibly due to the
entry of foreign banks.
Figure 3.1 Average and percentage change- input and output variables
3.4 Results
3.4.1 Efficiency results
This section presents the results of the application of the DEA meta-frontier
analysis to evaluate the efficiency of banks in selected countries in the
Mediterranean region.
Table 3.5 reports descriptive statistics of efficiency scores for the
countries under observation as well as estimates for all countries combined
(meta-frontier). Technical efficiencies and meta-technology ratios are
estimated for each country in each of 14 years of analysis, relative to a
balanced panel data set rather than relative to yearly frontiers, which
makes analysis of the evolution of efficiency over time meaningful.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 83
Table 3.5 Country-specific DEA efficiency scores (DEA-C)
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean
CY 96.9 97.2 96.6 97.9 100.0 98.2 99.8 100.0 97.3 95.2 95.6 95.1 92.9 99.3 97.3
DZ 100.0 94.4 93.1 95.7 100.0 100.0 100.0 100.0 94.7 90.4 91.3 90.2 91.4 98.6 95.7
EG 89.4 88.7 91.3 88.6 86.1 86.6 85.1 83.9 80.8 89.5 90.4 89.5 83.7 79.0 86.6
ES 90.4 90.3 87.4 86.2 83.1 87.4 87.3 86.4 87.1 84.1 83.0 85.2 86.1 87.5 86.5
GR 96.0 94.0 94.9 88.1 94.8 96.5 97.5 96.9 94.0 91.5 94.2 89.6 93.2 93.4 93.9
IL 94.9 93.0 98.2 93.4 96.5 96.5 99.0 96.6 89.7 98.6 99.1 93.8 95.3 88.9 95.2
IT 81.2 83.2 86.8 85.6 83.5 81.6 83.2 80.8 68.4 79.4 80.8 80.5 84.0 82.3 81.5
MA 100.0 97.3 98.0 96.6 99.2 96.7 100.0 100.0 100.0 98.2 100.0 100.0 100.0 100.0 99.0
MT 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 99.2 100.0 100.0 99.9
PT 89.8 89.4 91.6 95.4 95.1 91.2 93.7 93.5 90.3 89.1 93.4 92.7 93.5 95.3 92.4
TN 80.3 86.1 92.5 92.5 91.3 96.9 94.9 89.3 92.8 93.0 93.0 96.4 87.8 94.9 91.6
Mean 92.6 92.2 93.7 92.7 93.6 93.8 94.6 93.4 90.5 91.7 92.8 92.0 91.6 92.7
The average annual efficiency scores of banks of each country relative
to each country’s frontier (DEA-C) reveal a general steady trend. The
results relative to the country frontiers are reported only for completeness
of the analysis. Recall that these efficiencies are calculated relative to each
country’s frontier; the boundaries of these frontiers are restricted
technology sets, where the restrictions derive from the available economic
infrastructure and other characteristics of the production environment, as
discussed above. Also recall the small number of observations in some
countries, which causes dimensionality problems
We now move to the crucial part of this analysis, the measurement of
efficiency relative to a meta-frontier, defined as the boundary of an
unrestricted technology set. It is interesting to note that in most countries,
the country-specific frontiers were at least partially tangent to the metafrontier.
This is the case when at least one observation from each country
lies both on the country and on the meta-frontier and it is therefore
positioned in the point of tangency between the country and the metafrontier.
This indicates that the meta-frontier closely envelops the countryspecific
frontiers and that the value of the technological gap ratio equals the
maximum value of one for at least one observation in each of the sample
countries.
Looking at the efficiency scores derived from the estimation of the
meta-frontier as displayed in Table 3.6, Spanish banks dominate the region,
with average efficiency scores of 80.4%. Cypriot, Egyptian and Tunisian
banks are lagging behind with average efficiency scores of 55.6%, 49.5%

84 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
and 58.1% over the period. The region’s average efficiency score is 63.5%,
which indicates that Mediterranean banks could, on average, reduce costs
(inputs) by 36.5% and still produce the same outputs.
Table 3.6 Mean meta-frontier DEA efficiency scores
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean
CY 47.7 52.2 70.0 57.9 46.1 53.1 61.6 53.7 46.1 42.5 48.2 60.8 69.9 69.2 55.6
DZ 74.2 78.3 82.9 94.3 87.8 71.2 70.0 73.2 64.3 57.2 52.1 73.7 72.6 93.7 74.7
EG 51.7 51.4 60.1 56.1 43.2 45.3 49.3 49.8 44.5 45.7 45.4 53.3 52.0 45.7 49.5
ES 59.3 65.7 84.0 85.3 82.3 87.2 85.5 85.7 75.7 83.0 82.9 84.2 85.0 79.2 80.4
GR 45.4 43.5 52.7 48.4 41.0 48.4 65.3 62.4 56.6 50.4 52.9 54.2 63.9 68.1 53.8
IL 74.6 92.3 69.9 56.7 46.2 52.2 48.5 67.1 43.2 42.4 37.5 66.6 74.2 77.9 60.7
IT 44.5 47.9 58.6 65.6 63.9 69.8 71.8 71.8 57.6 63.2 62.0 64.0 70.3 73.5 63.2
MA 57.8 55.9 74.4 69.2 58.8 65.1 71.8 60.3 49.1 45.4 49.7 65.0 85.5 93.9 64.4
MT 76.1 79.5 87.3 81.5 63.4 65.2 67.7 57.3 63.8 56.2 59.7 62.9 62.7 70.4 68.1
PT 59.8 62.9 80.8 73.0 68.1 76.3 78.2 78.0 65.5 67.6 70.3 66.7 68.5 64.2 70.0
TN 43.3 46.9 82.5 76.3 62.9 70.1 72.3 63.4 47.8 48.2 44.1 45.2 49.9 60.6 58.1
Mean 57.7 61.5 73.0 69.5 60.3 64.0 67.4 65.7 55.8 54.7 55.0 63.3 68.6 72.4 63.5
Figure 3.2 illustrates the trend of efficiency levels over the period
1995-2008. For all countries, it is possible to note an improvement in
efficiency levels in the later stages of the analysis, from 2005 onwards (with
the exception of Egypt). This improvement is particularly remarkable for
Algerian and Moroccan banks. The overall mean efficiency in the region is
improving, once again driven by improvements in the best practice.
Minimum average efficiency scores also seem to be increasing, following a
drop in 2008.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 85
Figure 3.2 Average meta-frontier DEA scores

86 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
Table 3.7 illustrates the meta-technology ratios. The meta-technology
ratio (DEA-MTR) or technological gap, is calculated as the ratio between
meta-frontier (in)efficiency (DEA-M) and the country-specific (in)efficiency
(DEA-C) and it indicates the relative productivity of technologies. The
higher the ratio, the closer a country’s production technology is to the ‘best
practice’ in the region. Vice versa, the lower the ratio, the bigger is the
technology gap. An increase in the meta-technology ratio can be seen as
convergence towards the best practice.
Table 3.7 Average meta-technology ratios
1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 Mean
CY 48.9 53.5 72.5 59.1 46.1 54.0 61.7 53.7 47.1 44.6 50.3 63.6 75.2 69.6 57.1
DZ 74.2 84.6 90.4 98.7 87.8 71.2 70.0 73.2 68.8 65.2 57.7 85.9 80.8 95.1 78.8
EG 57.6 57.3 65.5 63.3 50.4 52.6 58.1 59.0 54.0 50.8 50.1 59.1 62.0 57.3 56.9
ES 65.3 72.3 96.0 99.0 98.9 99.8 97.7 99.1 86.5 98.4 99.8 98.7 98.8 90.4 92.9
GR 47.0 46.1 55.0 55.2 42.9 49.9 66.8 64.4 60.0 54.8 56.0 60.2 68.5 73.1 57.1
IL 78.2 99.2 71.2 60.3 47.8 54.1 49.1 69.6 48.1 43.1 37.8 70.4 77.6 86.8 63.8
IT 54.2 56.9 67.3 76.5 76.6 85.6 86.3 89.0 85.1 79.6 76.7 79.6 84.0 89.7 77.6
MA 57.8 57.2 75.8 71.6 59.2 67.1 71.8 60.3 49.1 46.2 49.7 65.0 85.5 93.9 65.0
MT 76.1 79.5 87.3 81.5 63.4 65.2 67.7 57.3 63.8 56.2 59.7 63.3 62.7 70.4 68.1
PT 66.2 70.6 87.8 76.4 71.2 82.9 83.3 83.3 72.3 75.8 75.0 71.7 73.4 67.5 75.5
TN 52.5 54.0 89.5 82.6 68.7 72.4 76.3 70.8 51.4 52.1 47.8 47.0 58.3 64.4 63.4
Mean 61.6 66.5 78.0 74.9 64.8 68.6 71.7 70.9 62.4 60.6 60.1 69.5 75.2 78.0 68.7
During this period of analysis, the average meta-technology ratio is
increasing indicating an ability of banks in all countries to appropriate the
best available technology. Spanish banks exhibit the highest metatechnology
ratio; the ratios also display increasing trend overtime. This
indicates that Spanish banks consistently improved their performance and
their technology became best practice. The dominance of Spanish banks is
clearly illustrated in Figure 3.3. Moroccan and Algerian banks seem to
catch up with best practice.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 87
Figure 3.3 Average meta-technology ratios
There is overall some indication of catching up with the best available
technology. However, the institutions comprising the sample are often the
largest banks in some countries and are for the majority foreign-owned in
the MENA region. As a result, the indication of catching up must be treated
with caution.

88 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
3.4.2 Convergence results
To evaluate β-convergence for our cross-section of Mediterranean
countries, we estimate equation (2) by OLS and GMM. Table 3.8 shows
regression estimates of the convergence coefficient β for the period 1995-
2008. The results from equation (2) that exclude the lagged dependent
variable are reported in the first column. The beta coefficient is always
negative and statistically significant, thus indicating that convergence in
efficiency scores has occurred across countries in the MED-11 area. The
results are confirmed in all three models although the goodness of fit for
the SYS-GMM (last column) shows that the p-value for AR(1) is greater
than 5%.
Table 3.8 Beta convergence
Coefficients
Equation (2) without
lagged dependent
variable
Pooled OLS
Robust
Pooled OLS
robust
β -.2502***
-.3174***
(.0588)
(.0679)
ρ - -.1682*
(.8813)
α 1.0481***
1.321***
(.2461)
(.2843)
Equation (2)
SYS-GMM
two step robust
-.3952***
(.1816)
+.1516
(.1801)
1.6448**
(.7520)
Goodness of fit:
R2 0.1227 0.1626
m1 p-value 0.245
m2 p-value 0.720
Sargan/Hansen 1.000
Note: OLS= Ordinary Least Squares; SYS-GMM= System GMM.
*,**,*** indicates significance at the 10%, 5% and 1% levels. Asymptotic standard error in
parentheses. Two-step estimates are Windmeijeier corrected (Windmeijer, 2005). m1 and m2
are tests for first-order and second-order serial correlation. Sargan & Hansen is a test of the
over-identifying restrictions for the GMM estimators.
Table 3.9 reports the results for the σ-convergence. In our case sigma
convergence indicates how quickly each country’s efficiency levels are
converging to the average. Recall that σ

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 89
convergence of towards ; the larger is σ in absolute value, the faster
the rate of convergence. We firstly estimated the model with pooled OLS
and fixed effects (the Hausman test allows us to reject random effects).
Potential problems with these two models are addressed by the estimation
of a dynamic GMM model. The last column of Table 3.10 reports the SYS-
GMM estimations results (equation 3). Following Arellano and Bover
(1995) and Blundell & Bond (1998), the use of a GMM estimator should
help mitigate possible endogeneity problems and omitted variable bias.
Results for all the estimations suggest an increase in the speed of
convergence as the σ coefficient is always negative and statistically
significant. Further, the SYS-GMM results satisfy the three additional
conditions: a significant AR(1) serial correlation, lack of AR(2) serial
correlation and a high Sargan/Hansen test.
Table 3.9 Sigma convergence (dependent variable ΔE)
Coefficients Equation (3) without
lagged dependent
variable
Pooled
OLS
robust
Fixed
effects
Pooled
OLS
robust
σ -.1874*** -.4131*** -.2129***
(.0607) (.0687) (.0641)
ρ -.0151
(.0972)
µ -.00032 -.0074 -.0029
(.0093) (.0100) (.0097)
Equation (3)
Fixed
effects
-.5645***
(.0805)
-.1271
(.0864)
-.0095
(.0092)
SYS-GMM
two step
robust
-.1791***
(.4521)
-.1359
(.4760)
-.0019
(.0198)
Goodness of fit:
R2 0.0961 0.1194
F-test 39.09*** 26.08***
m1 p-value 0.342
m2 p-value 0.843
Sargan/Hansen 1.000
Note: OLS= Ordinary Least Squares; SYS-GMM= System GMM.
*,**,*** indicates significance at the 10%, 5% and 1% levels. Asymptotic standard error in
parentheses. Two-step estimates are Windmeijeier corrected (Windmeijer, 2005). m1 and m2
are tests for first-order and second-order serial correlation. Sargan &Hansen is a test of the
over-identifying restrictions for the GMM estimators.

90 | ANALYSIS OF EFFICIENCY AND CONVERGENCE
3.5 Conclusions
This study examines the dynamics of cost efficiency in 11 Mediterranean
countries over the period 1994-2008. For all countries, the results indicate
an improvement in efficiency levels in the later stages of the analysis, from
2005 onwards (with the exception of Egypt). This improvement is
particularly remarkable for Algerian and Moroccan banks. The overall
mean efficiency in the region is improving, once more driven by
improvements in the best practice. Spanish banks dominate the region,
with average efficiency scores of 80.4% against the region's average of
63.5%. Spanish banks also exhibit the highest meta-technology ratio and the
ratios increase overtime. This indicates that Spanish banks consistently
improved their performance and their banking technology became best
practice. Nonetheless, during this period of analysis, the average metatechnology
ratio is increasing, indicating an ability of banks in all countries
to appropriate the best available technology.
These results are supported by the estimation of β-convergence. The β
coefficient is always negative and statistically significant, thus indicating
that convergence in efficiency scores has occurred across countries in the
MED-11 area. Furthermore, results for the σ-convergence suggest an
increase in the speed of convergence as the σ coefficient is always negative
and statistically significant. This indicates that, whereas the technological
gap is still wide, the gap is narrowing at a faster speed.

4. IMPACT OF BANK REGULATIONS ON
104 |
EFFICIENCY
A
lthough the quality and adequacy of banking regulation and
supervision are often touted as the essential factors contributing to
a sound and well-performing banking sector, few studies have
produced empirical evidence to back these assertions. A common
finding is that certain specific regulatory elements may have a positive
impact, while others may do the opposite or invite instability. Moreover, an
adequate and well-functioning regulatory system appears to improve
various performance or stability measures as long as they are
complemented by other institutional and macroeconomic conditions.
This section focuses on a very specific question: Are the banks in the
Mediterranean more cost efficient in countries with sounder regulatory and
supervisory conditions? The results echo the recent findings in the
literature. Certain regulatory aspects, such as disclosure requirements,
credit information availability and entry obstacles, are highly important.
The presence of an explicit deposit insurance scheme also improves
efficiency, drawing attention to the importance of enhancing confidence for
depositors. Other findings are less clear and require further investigation.
For example, although restrictions on activities lower efficiency, it is
possible that they could lead to increased risks.
The next section provides an overview of the literature. Then, section
4.2 summarises the data sources and gives an extensive summary of the
variables used and their hypothesised impacts. Section 4.3 discusses the
empirical results and section 4.4 summarises the main findings.
4.1 Literature review
The literature on the regulatory and supervisory determinants of bank
efficiency is still in its infancy. Most studies use relatively broad measures

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
of regulatory and institutional conditions. For example, Dietsch & Lozano-
Vivas (2000) and Bos & Kool (2006) draw attention to the importance of a
number of “environmental factors”, including those relating to bank
structure and regulation, such as concentration ratios, capital strength and
intermediation ratios. The authors find that banks that operate in less
concentrated markets, with greater capital and higher intermediation ratios
tend to have lower costs. Similarly, Fries & Taci (2005) focus on transition
economies using a broad measure of banking sector reforms, developed by
the European Bank for Reconstruction and Development (EBRD), to find
that banks in countries with an active agenda tend to perform better and
have higher profitability.
Our study is similar to several recent studies that assess empirically
the impact of regulations on different measures of bank efficiency. Barth et
al. (2006) use the BRSS database for the years 2000 and 2003 to identify the
regulatory and institutional determinants of net interest margins and cost
efficiency in 68 countries. The authors’ results provide partial support for
the importance of capital regulations and supervisory power. More
specifically, aside from private monitoring, most of the variance in interest
margins and overhead costs are explained by institutional and macroeconomic
factors. The stringency of capital requirements and the power
bestowed on supervisory authorities are at best weakly associated with
greater efficiency. These findings provide a broad support for the third
pillar of Basel II.
More recently, Pasiouras (2008) also uses the BRSS databases and a
large sample of banks from 95 countries but develops cost and scale
efficiency measures. Although the results support all three pillars of Basel
II, the results are especially strong for the market discipline mechanisms
(i.e. the third pillar). In addition, the role of bank-specific factors, such as
bank liquidity and capitalisation, as well as market-specific factors, such as
access to banking and presence of government-owned banks, are
reaffirmed in the study.
In a later study, Pasiouras et al. (2009) assess the impact of regulatory
conditions on profit and cost efficiency of banks using similar data. The
authors find that regulations that improve supervisory power and market
discipline tend to have a positive impact on both of the measures. In turn,
capital requirements tend to improve only cost efficiency while reducing
profit efficiency. In addition, the results show that restricting banks’
activities may improve their profit efficiencies and worsen their cost
efficiencies.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 93
4.2 Methodology and data
The model for assessing the impact of regulatory and institutional factors
on bank efficiency is as follows:
where B stands for bank-specific factors that relate to bank i at time t
while C and R stand for country-specific and regulatory factors and MTR is
the meta-technology ratio as derived in section 3 which is the bank-specific
meta-frontier efficiency divided by the country-specific efficiency.
The Barth et al. database provides up to three observations for each
country, based on the 2000, 2003 and 2007 surveys. Due to different
completion times, the surveys give a glimpse of the regulatory conditions
between the publication year and one (or even two) years prior. For
example, the 2000 survey was sent to the authorities in 1998, with most
results arriving in 1999-2000. In order to minimise potential errors from
misalignments between regulatory and non-regulatory factors, the later
were averaged over a relevant time period. Moreover, the time spans were
chosen to ensure that the current explanatory variables are used to explain
future efficiency scores. Table 4.1 details the correspondence between the
regulatory and non-regulatory variables.
Table 4.1 Sample correspondence for survey years
Survey year Meta-tech. efficiency
(MTR)
Other variables
(B, I, M)
2000 Avg. of 1999-2001 Avg. of 1998-2000
2003 Avg. of 2002-04 Avg. of 2001-03
2007 Avg. of 2006-08 Avg. of 2005-07
Note: For all non-regulatory variables (i.e. MTR, B, I, M), the averages for the given periods
were used.
The level of coverage of the sampled banks is depicted Table 4.2.
When the entire sample is considered, the database covers just over the half
of the entire banking assets. The coverage in the South-MED is significantly
more partial, with the total assets of the banks in the sample representing
between one-quarter to one-third of the total assets of the banks in the
region. Moreover, the total activities of the South-MED account for a small
proportion of the entire sample. More specifically, the total assets of the
South-MED banks within the sample represent are between 5 to 10% of the

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
total assets of all banks in the sample. This is simply an outcome of the size
of the EU’s banking market. In order to ensure a balanced database, Italy’s
larger banks were excluded.
Table 4.2 Coverage of sample
Banks in sample
(number of banks)
Coverage
(% of bank assets in
country)
Share in sample
(% of entire sample assets)
2000 2003 2007 2000 2003 2007 2000 2003 2007
Algeria 2 2 2 27.6% 10.3% 8.8% 0.48% 0.97% 0.89%
Egypt 14 14 14 32.4% 34.5% 47.8% 3.10% 1.91% 1.58%
Israel 5 5 5 13.6% 12.6% 15.6% 5.05% 3.35% 2.33%
Morocco 4 4 4 35.1% 38.7% 54.0% 0.96% 0.83% 0.83%
Tunisia 10 10 10 73.1% 72.2% 74.3% 0.49% 0.42% 0.31%
SOUTH-
MED
35 35 35 24.5% 23.9% 31.5% 10.51% 7.65% 5.96%
Cyprus 4 4 4 51.4% 73.1% 79.9% 1.09% 0.91% 1.19%
Spain 65 65 65 97.8% 94.1% 95.0% 29.08% 32.66% 38.31%
Greece 8 8 8 65.9% 66.5% 69.2% 4.96% 4.63% 4.99%
Italy 80 80 80 19.8% 19.2% 17.4% 45.79% 46.19% 43.34%
Malta 4 4 4 46.6% 49.2% 30.7% 0.41% 0.39% 0.49%
Portugal 10 10 10 59.8% 64.4% 72.4% 8.16% 7.58% 5.73%
EU-MED 171 171 171 55.1% 52.4% 60.0% 89.49% 92.35% 94.04%
ENTIRE
SAMPLE
206 206 206 51.9% 50.3% 58.3% 100.00% 100.00% 100.00%
Sources: Bankscope, national central banks and the ECB.
Among the 11 countries in the sample, Algeria has the lowest
coverage. This is entirely due to the fact that most of the Algerian banks are
publicly-owned for which little information exists. All of the covered banks
in Algeria are owned privately. For Israel, the detailed balance sheet
information was available for only a small share of the banks. In other
South-MED countries, the coverage is more complete. This is particularly
the case in Tunisia where the total assets of the sampled banks represent
nearly three-quarters of the total assets of all banks in Tunisia.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 95
4.2.1 Bank-specific variables
Four bank-specific variables are used to control for the market power, size,
liquidity and capital strength.
The natural logarithm of a bank’s assets, defined as the bank assets
variable, serves as an indicator of the bank’s size. Size could be a
determinant of costs if there are increasing returns to scale. For example,
larger banks may be able to reduce their operating costs by cutting back on
personnel and administrative costs. Moreover, if a fixed cost is associated
with financial transactions, larger banks may also be able to recue such
costs. Lastly, larger banks may be in a better position to diversify their risks
and thus reduce their borrowing costs. 68
Market power is measured by the bank market share, i.e. the share of
a bank’s assets in total banking assets for the relevant years. According to
the traditional ‘quiet life hypothesis’, banks that are in a dominant position
are unlikely to occupy themselves with cost reduction and are likely to
behave inefficiently (Hicks, 1935). Alternatively, managers of banks with
extensive market power may have other incentives than being efficient,
such as ‘building empires’, (Hughes et al., 2003). These theories would
suggest that market share would be negatively correlated with efficiency.
Bank liquidity will be measured by the ratio of cash and due from
central bank and other banks (i.e. demand and time deposits maintained in
other banks) to customer deposits. Capital strength is measured by share of
bank equity in total assets. Well-capitalised and liquid banks tend to face
lower default risks and are thus likely to face lower funding costs. On the
other hand, in the Middle East and Northern Africa (MENA) region, such
banks tend to hold significant amounts of government debt, possibly under
direct or indirect government control and with extensive market power,
both of which could lead to inefficiencies. Thus, there is no clear
relationship between these two variables and bank efficiency.
68 The literature has obtained mixed results on the impact of bank size or scale on
efficiency, although most studies have found that large banks are either more
efficient or equally efficient as smaller banks. See DeYoung (1998) for a general
discussion. For results that are applicable to the Middle East and the North Africa
(MENA) region, see Olson & Zoubi (2010), who provide evidence for scale
economies.

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
4.2.2 Country-specific variables
Three country-specific variables are considered:
Inflation is often thought to increase instability and decrease bank
efficiency, since it makes price discovery harder and makes interest rates
less informative about the underlying conditions. High inflation may
increase labour costs and, by increasing the number of transactions, lead to
an increased competition in excessive branching and other operational
costs. Additionally, inflation exacerbates information asymmetries,
increasing the costs of state verification (Huybens & Smith, 1999). Economic
growth, or more specifically real GDP growth, is included to control for
business cycles.
The third indicator, institutional quality, is built by aggregating
eight dimensions of the quality of political institutions. These dimensions
are: i) polity, which measures the relative strength of democratic (or
conversely autocratic) institutions, determining the extent to which the
executive arm is controlled by regular checks and balances, guarantee of
civil liberties, freedom of political expression and participation; ii)
executive openness, which measures the openness of the executive
recruitment; iii) executive competitiveness, which controls for the
competitiveness of the election procedures; iv) executive constraints,
measuring the extent of authority that can be practiced by the executive
arm; v) political competitiveness, which measures the competitiveness of
the political arena; vi) control of corruption, concerning the perception that
public authorities can exercise their power without obtaining private gain;
vii) voice and accountability, which measure the degree to which citizens
can voice their opinions and desires in the political system; and viii) rule of
law, capturing the perceptions on the quality of contract enforcement,
property rights, police, courts and likelihood of crime and violence. The
institutional quality variable is the first principal component of the seven
variables identified above. By using a single variable to account for the
various dimensions, the principal component analysis effectively addresses
the potential multi-collinearity concerns that would arise from including
the variables collectively.
Institutional and political conditions are likely to be very important in
determining bank behaviour, stability and performance. For example,
giving authorities extensive powers in countries where political freedoms
and checks and balances are limited could lead to a misuse of authority.
Indeed, Barth et al. (1999) and La Porta et al. (2002), among others, find that
prevalence of state is associated with poorly operating financial systems.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 97
Public interference may also be an instrument for politicians to expand or
maintain their power (Shleifer & Vishny, 1994). Another interesting
question is whether banking stability closely reflects the ambitions of
politicians who are in power. 69
A number of studies find that institutional conditions matter in the
determination of bank efficiency. Pasiouras (2008) find that the degree to
which a country’s laws protect private rights matters substantially, such
that banks in better-governed countries are much more efficient. Using a
sample that is similar to the present study, Ben Naceur et al. (2009) show
that several institutional factors, most notably the quality of the judicial
system and a better legal system, are crucial in explaining cross-country
differences in bank efficiency across four Middle East and North Africa
countries: Egypt, Jordan, Morocco, and Tunisia.
4.2.3 Regulatory variables
The regulatory factors are mostly based on the Barth et al. surveys (BRSS)
and, for the case of credit information, on World Bank’s Doing Business
Surveys. The variables used and a brief description of their construction
methodology, already detailed in section 2, are presented below.
The index scope restrictions measures the degree of restrictions on
what a bank can do, including prohibitions on key non-traditional
activities, such as securities and insurance underwriting, brokering,
dealing, real estate development and so forth. 70 In theory, restricting the
range of activities may have opposing effects on efficiency. On the one
hand, allowing banks to take equity positions can exacerbate the moral
hazard problems between a borrower and a lender, adversely affecting the
optimality of investment decisions and the overall bank efficiency (Boyd et
al., 1998). On the other hand, when banks engage in a broader set of
activities, their ability to diversify risks is also enhanced. A more stable
69 Using data on 25 emerging countries, Brown & Dinc (2005) find that bank
failures are significantly less likely to occur prior to an election, pointing to a
concern for loss of votes and increased attention paid to stability. Indeed, among
the failures observed in the sampled countries in the authors’ dataset, only 10%
have taken place within a year before the election. The study also raises another
interesting point on crisis management: banks that are taken over by the
government almost never fail, reflecting the underlying guarantees.
70 For more details on the scope restrictions index, see section 2.2.

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
income stream can serve to reduce borrowing costs, which will increase
efficiency.
Entry obstacles are measured by aggregating the degree of licensing
restrictions, the rate of foreign denials (share of denials in total applications
over the past five years) and the total market share of the governmentowned
banks in terms of total assets. A regulatory structure that is more
amenable to entry is likely to enhance competitive conditions, possibly
undermining market power and the potential for a ‘quiet life’ (see above).
Several studies find that foreign entry is associated with more competitive
conditions, translating into lower costs and profits for domestic banks,
(Claessens et al., 2001; Claessens & Laeven, 2004). Highly predominant
state-ownership of the banking sector may also undermine the competitive
conditions. 71
The capital requirement stringency considers whether there are
explicit requirements on the amount and type of capital allowed, risk
adjustments and initial capital. 72 Although stricter rules on capital and an
autonomous supervisor may make the system as a whole sounder, the
impact on efficiency is less than clear. More and better capital held by one
bank could translate into lower borrowing costs and may signal operational
efficiency. However, when the conditions apply to all the banks, the
informational benefits do not materialise. Moreover, more stringent rules
may increase compliance costs, undermining efficiency.
The index of supervisory independence measures the autonomy of
the supervisor from political influence, considering whether the supervisor
is ultimately accountable to a minister, could be sued for his/her actions
committed in exercise of his/her duties, and whether the head of the
agency has an undetermined (i.e. non-fixed) term. 73 Under the ‘private
interest view’ to regulation, politically-oriented regulators fail to maximise
social welfare and may thus undermine private-sector efficiency (Shleifer &
Vishny, 1994; Barth et al., 2006).
71 Although public banks may have a development role (Gerschenkron, 1962;
Stiglitz, 1994; Hakenes & Schnabel, 2006) and may even be more stable than their
commercial peers (Garcia-Marco & Robles-Fernandez, 2008; Ayadi et al., 2009),
there is a general agreement that is backed by substantial evidence that undue
public interference leads to an inefficient allocation of credit and risk-taking
(Sapienza, 2004; Dinc, 2005; Khwaja & Mian, 2005; Cole, 2009).
72 For more details on the capital requirement stringency index, see section 2.4.
73 For more details on the supervisory independence index, see section 2.5.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 99
The deposit insurance variable is a dummy that identifies the
existence of an explicit deposit insurance scheme. Considering the index’s
impact on bank efficiency, there may be opposing forces at play. Several
studies have noted that excessive guarantees may result in moral hazard
and may encourage excessive risk-taking and increase the likelihood of
crises (Merton, 1977; Bhattacharya & Thakor, 1993; Demirgüç-Kunt &
Detragiache, 2002; Demirgüç-Kunt & Kane, 2002). These risks could
increase borrowing costs, thereby lowering efficiency. However, having a
safety net could also enhance efficiency, especially in the context of
developing countries where the (shadow) cost of funds is often high since
many potential depositors may not open an account or due to the risks of
bank runs (Diamond & Dybvig, 1983). By reinforcing the soundness of
deposits, the schemes may thus lower costs and enhance efficiency.
The private monitoring variable is an indicator for the disclosure of
information, which is at the core of market discipline. 74 The disclosure of
reliable and timely information allows investors and depositors to better
understand and monitor the underlying risks and inefficiencies and can
serve as a disciplining tool on the bank’s management. 75 There are some
questions, however, on whether disclosure requirements and practices can
really function in countries with poor accounting standards and
underdeveloped capital markets, which tend to be the primary customers
of such information. In responding to these concerns, Caprio & Honohan
(2004) note that despite these shortcomings, market discipline could work
to discipline banks, especially in countries with no credible deposit
guarantees—explicit or implicit—where market participants have strong
incentives to engage in monitoring.
Credit information is an indicator for the availability, coverage and
depth of credit information. 76 Information-sharing can impact efficiency
levels through various channels. First, it can reduce market power by
breaking the information monopolies developed by existing banks (Vives,
1990). In this manner, making information more available can work against
the ‘quiet life’ that the incumbent banks enjoy. Second, it improves the
74 For more details on the private monitoring index, see section 2.7.
75 The idea is at the core of the third pillar of the Basel II framework. Basel
Committee on Banking Supervision’s (1999) consultative paper on capital
adequacy asserts that “[m]arket discipline imposes strong incentives on banks to
conduct their business in a safe, sound and efficient manner” (BCBS, 2001, p. 1).
76 For more details on the strength of credit information index, see section 2.8.

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
accuracy of credit-worthiness assessments of banks, thereby reducing
credit risks and the efficiency in the allocation of credit (Pagano & Jappelli,
1993).
4.2.4 Data sources
Perhaps the most important challenge that researchers face in attempting to
assess the impact of regulations is the availability of reliable data on
regulatory conditions. Presently, the standard dataset that quantifies the
quality and adequacy of banking regulations for a large set of countries
over time is based on the detailed results of the Financial Sector
Assessment Program (FSAP). Undertaken jointly by the IMF and the World
Bank since 1999, the FSAP regularly evaluates the regulatory structures of
its members by assessing their compliance with international standards.
For the banking regulations, the assessors use the so-called ‘Basel Core
Principles’ (BCPs) of the Basel Committee on Bank Supervision, which
were issued in 1997 as a basis for their evaluations. 77 These assessments are
conducted according to standardised methods developed by the Basel
Committee and result in a score of compliance on each one of the 25 BCPs. 78
The IMF has compiled an in-house database that provides the level of
compliance in each one of BCPs for all the evaluated countries since 1999. 79
As the earliest study of its kind, Sundararajan et al. (2001) have used
the BCP database to show that it does a poor job in explaining interest
spreads and credit risk. Attempting to explain this counter-intuitive result,
77 The Basel Committee on Bank Supervision comprises representatives from bank
supervisory agencies from advanced countries, including Australia, Belgium,
Canada, France, Germany, Italy, Japan, Luxembourg, the Netherlands, Spain,
Sweden, Switzerland, the United Kingdom and the United States, as well as
developing countries, including Argentina, Brazil, China, Hong Kong SAR, India,
Indonesia, Korea, Mexico, Russia, Saudi Arabia, Singapore, South Africa and
Turkey.
78 For a thorough review of these methodologies, see World Bank (2005).
79 The publication of the FSAP results, including the detailed Reports on the
Observance of Standards and Codes (ROSCs), is voluntary. Although most
developed countries have agreed to publish the detailed assessments and ROSCs,
among the South-MED countries in our sample, a detailed account of compliance
with BCPs is only available for Tunisia. Moreover, Egypt has agreed only to
publish a summary of the FSAP 2008 report. For these reasons, the compilation of
the BCP compliance scores was not possible for our sample.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 101
Das et al. (2005) find that regulatory quality leads to a more sound banking
sector with less liquidity stresses as long as cross-country differences in
institutional quality are accounted for. Pointing at a stronger degree of
conditionality, Podpiera (2006) finds that a greater compliance with the
BCPs has enhanced asset quality and bank performance when various
financial, macroeconomic and structural factors are controlled for. More
recently, Demirgüç-Kunt & Detragiache (2010) fail to find a relationship
between BCP compliance and systemic risk measures.
Taking a different route, Barth et al. (2004) have compiled an
alternative dataset on banking laws and regulations, also used in this
study. The authors’ dataset (referred henceforth as the Barth et al.
regulatory and supervisory survey or the “BRSS”) is based on the results of
a worldwide survey, collected from national regulatory authorities in over
150 countries in the years 2000, 2003 and 2007. 80 Analysing the data, the
authors find regulatory systems that facilitate adequate private monitoring
(i.e. disclosure requirements) tend to have a beneficial effect across almost
all the indicators they consider. Other regulatory variables, such as the
official supervisory power or capital requirements, have no or little impact.
In later work, Barth et al. (2006) show that several institutional factors, such
as the absence of corruption and the presence of voice and accountability,
also have a strong positive impact on net interest margins and overhead
costs. Using the BRSS, Laeven & Levine (2008) find that the impact of
regulations on risk depends crucially on a bank’s ownership structure. In
particular, banks with more powerful shareholders tend to take more risks,
making the link between regulatory environment and risk-taking
ambiguous (even negative) when ownership is not accounted for.
Comparing the different approaches to measuring regulatory
adequacy and quality, one of the main questions is whether the BRSS and
the IMF’s BCP assessments are measuring the same things. In theory, there
should be substantial correlation between the two databases since both are
based on comparable principles. However, Čihák & Tieman (2008) note that
the correlation between the two databases is exceptionally low. The
differences can arise due to a variety of reasons. Neither measure is perfect
in fully implementing the laws as “countries may change the regulatory
framework to appear better on paper but not on ground” (Barth et al., 2006,
pp. 81-82). The BCP assessments are likely to be more illuminating in this
80 The covered years are approximate as the responses have been collected over
several years in each one of the countries.

128 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
manner, as they depend on independent assessments and not on selfevaluations.
The BRSS, which relies on answers to standardised questions,
is less likely to contain potential ‘grading biases’ that are more likely in the
BCP assessments.
Table 4.3 details the data sources and provides a descriptive
summary of the variables used in the regressions in this section. The
institutional quality and the regulatory factors exhibit substantial variance,
which could make them helpful in explaining the variability in the bank
efficiency scores.
Table 4.3. Data sources and descriptive statistics
Variable name Source Obs. Mean St. dev. Min Max
Bank efficiency Own calculations & 618 0.817 0.160 0.310 1.074
(MTR)
Bankscope
Bank assets Bankscope 618 8.165 1.523 3.456 13.724
Bank market share Bankscope 618 0.047 0.111 0.000 0.780
Bank liquidity Bankscope 618 0.033 0.044 0.000 0.443
Bank equity Bankscope 618 0.088 0.045 -0.037 0.364
Inflation WDI 618 2.754 0.874 1.317 7.278
Growth WDI 618 2.944 1.579 0.230 6.134
Institutional quality Polity IV &
Kaufmann et al.
(2009)
606 9.777 3.726 -0.060 11.699
Scope restrictions Barth et al. surveys 596 7.065 1.528 5.000 10.000
Entry obstacles Barth et al. surveys 596 8.155 11.556 0.292 55.233
Cap. req. stringency Barth et al. surveys 596 5.045 2.863 1.000 9.000
Supervisory indep. Barth et al. surveys 596 1.099 1.027 0.000 3.000
Deposit insurance Barth et al. surveys 596 0.904 0.294 0.000 1.000
Private monitoring Barth et al. surveys 596 8.440 0.995 6.000 10.000
Credit information Doing Business
surveys
400 4.705 1.256 0.000 6.000
Note: The Bankscope database is compiled and distributed by Bureau van Dijk; World
Development Indicators (WDI) and Doing Business surveys are both distributed by the
World Bank; Polity IV is developed and distributed by the Center for Systemic Peace and
Colorado State University.
4.3 Empirical results
This section investigates the impact of complying with the regulatory
standards developed in earlier chapters on efficiencies of banks. Since the
efficiency scores developed in section 3 are always positive and almost
always fall within the unit range, the dependent variable (i.e. the MTR) is a

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 103
limited dependent variable. The empirical estimations in this section use
the Tobit model. Two specifications were used to ensure that the results are
robust. First, Table 4.4 gives the results of a pooled regression. Second,
Table 4.5 provides results for random-effects panel regressions. The effects
of each of the regulatory variables were estimated separately in order to
reduce the potential multi-collinearity that exists between these variables.
Starting with the bank-specific variables, the results are mostly in line
with prior literature. Bank size, or the natural log of bank assets, has a
significant and robust impact on efficiency. Larger banks are more efficient,
which would confirm the presence of scale economies, in line with the
findings of Olson & Zoubi (2010). In turn, market power, as measured by
the market share, has generally a negative impact on efficiency. This result
confirms the ‘quiet life’ hypothesis, which suggests that the market power
that banks enjoy leads them to forego revenues or cost-saving
opportunities. Several studies have found evidence for such a relationship
with a variety of efficiency and performance measures, including Berger &
Hannan (1998) for the US banks and Maudos & de Guevara (2007) for the
EU banks. 81
Banks that are more liquid are significantly less efficient. As noted
repeatedly in earlier sections, the South-MED banks are more liquid
because they hold more government assets. Since the opportunity cost of
holding liquid assets are low, such banks are also more likely to hold more
cash and cash-like deposits in the central bank and other banks. The
relative inefficiency of these banks appear to confirm the ‘lazy banks’ view,
which suggests that banks in developing countries that invest in public
assets develop more slowly and are substantially less efficient (Hauner,
2008; Hauner, 2009).
The results on the impact of capitalisation on efficiency are less conclusive
for the strength of capitalisation. According to the pooled regression results
(
81 An alternative explanation is that the management of larger banks are more
interested in ‘building empires’. For a study documenting how such incentives
may give rise to lower bank efficiency, see Hughes et al. (2003).

Table 4.4), well-capitalised banks have a slightly higher efficiency than
other banks; however, these results are not robust and disappear
completely in the panel regressions (Table 4.5). It is entirely possible that
the positive impact of the strength of capitalisation is offset by the business
models of such banks, which, much like the highly-liquid institutions,
invest more in government assets.
Among the two macroeconomic factors, inflation clearly has the more
robust impact. More specifically, a lower rate of inflation contributes
substantially to greater bank efficiency, which is in line with the literature,
(Demirgüç-Kunt et al., 2004; Kasman & Yildirim, 2006). In turn, economic
growth is generally positively related to efficiency, pointing to economic
spillovers and potential opportunities for banks to reduce costs during
bursts of growth.
The institutional quality variable, which is an aggregation of a
number of institutional factors, including the strength of democratic
processes, political openness, power exercised by the executive arm,
election procedures, control of corruption, voice and accountability, and the
rule of law. The coefficient estimates are highly significant in all of the
specifications and point to a strong positive correlation between the quality
of institutions and efficiency. These findings are in line with recent findings
and highlight the fact that bank regulations cannot be viewed in isolation
from the overall institutional framework, (Demirgüç-Kunt et al., 2004; Barth
et al., 2006).
104 |

Table 4.4 Determinants of bank efficiency, pooled regressions
CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 105
I II III IV V VI VII
Bank assets 0.028*** 0.027*** 0.023*** 0.032*** 0.031*** 0.027*** 0.022***
(0.004) (0.005) (0.004) (0.004) (0.004) (0.004) (0.005)
Bank market
share
-0.408*** -0.460*** -0.385*** -0.486*** -0.502*** -0.424*** -0.188**
(0.053) (0.063) (0.055) (0.058) (0.054) (0.058) (0.080)
Bank liquidity -0.410*** -0.151 -0.486*** -0.416*** -0.398*** -0.475*** 0.138
(0.104) (0.119) (0.107) (0.115) (0.107) (0.112) (0.122)
Bank equity 0.191* 0.224 0.210* 0.326*** 0.262** 0.331*** 0.168
(0.116) (0.138) (0.119) (0.126) (0.120) (0.123) (0.135)
Inflation -0.041*** -0.004 -0.029*** -0.025*** -0.034*** -0.026*** -0.030***
(0.006) (0.007) (0.006) (0.006) (0.006) (0.006) (0.007)
Growth -0.001 0.010* -0.002 0.024*** 0.023*** 0.013*** 0.031***
(0.004) (0.005) (0.004) (0.004) (0.004) (0.004) (0.005)
Inst. quality 0.012*** 0.005 0.016*** 0.022*** 0.013*** 0.018*** 0.015***
(0.002) (0.004) (0.002) (0.002) (0.002) (0.002) (0.003)
Scope restrictions -0.042***
(0.004)
.. .. .. .. .. ..
Entry obstacles .. -0.006***
(0.001)
.. .. .. .. ..
Cap. req.
stringency
.. .. 0.020***
(0.002)
.. .. .. ..

106 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
Supervisory
indep.
.. .. .. 0.010*
(0.005)
.. .. ..
Deposit insurance .. .. .. .. 0.173***
(0.020)
.. ..
Private
monitoring
.. .. .. .. .. 0.032***
(0.006)
Credit
information
.. .. .. .. .. .. 0.049***
(0.009)
Constant 0.899*** 0.591*** 0.473*** 0.329*** 0.317*** 0.191*** 0.254***
(0.064) (0.063) (0.042) (0.042) (0.039) (0.049) (0.047)
Observations 584 380 584 584 584 584 400
Wald χ² (8) 465.4 284.5 437.3 360.3 424.4 386.0 299.9
Log likelihood 482.9 335.2 468.9 430.4 462.5 443.3 337.0
Notes: Standard errors in parentheses. The regressions use the Tobit estimation procedures to account for the limited dependent variable,
the meta-technology ratio (MTR), or the ratio of bank-specific meta-frontier efficiency and the country efficiency.
***, **, * represent statistical significance at the 1%, 5%, and 10% levels (p-values), respectively.

Table 4.5 Determinants of bank efficiency, random-effects panel regressions
CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 107
I II III IV V VI VII
Bank assets 0.028*** 0.022*** 0.024*** 0.026*** 0.027*** 0.020*** 0.023***
(0.005) (0.006) (0.005) (0.005) (0.005) (0.006) (0.005)
Bank market share -0.274*** -0.273*** -0.263*** -0.245*** -0.268*** -0.201** -0.120
(0.077) (0.085) (0.077) (0.080) (0.077) (0.081) (0.085)
Bank liquidity -0.306*** -0.184* -0.317*** -0.325*** -0.325*** -0.338*** -0.022
(0.095) (0.097) (0.095) (0.096) (0.093) (0.095) (0.106)
Bank equity -0.079 0.204 -0.108 -0.086 -0.070 -0.074 0.157
(0.127) (0.144) (0.128) (0.128) (0.125) (0.128) (0.132)
Inflation -0.016*** -0.012** -0.012*** -0.012*** -0.030*** -0.011*** -0.021***
(0.004) (0.005) (0.004) (0.004) (0.005) (0.004) (0.006)
Growth 0.009*** 0.011** 0.010*** 0.013*** 0.008*** 0.011*** 0.016***
(0.003) (0.005) (0.003) (0.003) (0.003) (0.003) (0.005)
Inst. quality 0.021*** 0.013*** 0.022*** 0.023*** 0.013*** 0.023*** 0.016***
(0.002) (0.004) (0.002) (0.002) (0.003) (0.002) (0.003)
Scope restrictions -0.010*** .. .. .. .. .. ..
(0.003)
Entry obstacles .. -0.003*** .. .. .. .. ..
(0.001)
Cap. req. .. .. 0.006** .. .. .. ..

108 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
stringency
Supervisory indep. .. ..
(0.002)
.. 0.003
(0.003)
.. .. ..
Deposit insurance .. .. .. .. 0.140***
(0.025)
.. ..
Private monitoring .. .. .. .. .. 0.012***
(0.004)
..
Credit information .. .. .. .. .. .. 0.035***
(0.008)
Constant 0.505*** 0.544*** 0.408*** 0.400*** 0.436*** 0.358*** 0.320***
(0.059) (0.072) (0.047) (0.048) (0.047) (0.050) (0.048)
Observations 584 380 584 584 584 584 400
Wald χ² (8) 258.2 226.9 252.8 229.8 282.5 243.6 278.8
Log likelihood 570.4 370.6 568.7 566.4 580.6 570.3 387.0
Notes: Standard errors in parentheses. The regressions use the Tobit estimation procedures to account for the limited dependent variable,
the meta-technology ratio (MTR), or the ratio of bank-specific meta-frontier efficiency and the country efficiency.
***, **, * represent statistical significance at the 1%, 5%, and 10% levels (p-values), respectively.

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 109
Turning to the regulatory factors, it is notable that several variables have
robust and significant impact. The presence of explicit deposit schemes,
greater disclosure practices for better private monitoring, and the
availability and use of credit information all contribute strongly to greater
efficiency. The same can also be said for the stringency of capital
requirements, although the impact is less significant in the panel regression
(column II) in Table 4.5. Restrictions placed on banking activities and entry
obstacles adversely affect bank efficiency. Prohibiting security and
insurance transactions may indeed reduce the banks’ ability to diversify
risks and activities which is by and large compatible with the literature on
the higher profitability of banking conglomerates and universal banking,
(Vander Vennet, 2002). Supervisory independence has only a weak impact
on efficiency in the pooled regression (column III of Table 4.4), possibly due
to the fact that other political factors, i.e. power of executive arm, are
readily controlled.
The positive impact of the availability of information on bank
efficiency is in line with the literature. In particular, the idea that disclosure
laws and practices that facilitate private monitoring tend to reduce costs is
echoed in several studies, including most notably Barth et al. (2006),
Pasiouras (2008) and Pasiouras et al. (2009). Moreover, Brown et al. (2009)
and Djankov et al. (2007) show that availability of credit information is
associated with lower transaction costs and moderation of credit risks,
enhancing the access of credit of opaque borrowers such as small- and
medium-sized enterprises (SMEs).
A more surprising result is the pro-efficiency impact of the presence
of deposit insurance schemes. The results reviewed in this section show
that Mediterranean banks operating under deposit insurance schemes are
significantly more efficient than those without such schemes. There is some
weak support for these findings in recent literature. For example, Pasiouras
(2008) finds that deposit insurance schemes improve efficiency (albeit at a
marginal level of significance) when other regulatory aspects are
considered alongside. In other cases, the availability of such schemes seems
to have no or negative impact on efficiency. The results obtained here
would seem to support the idea that having a safety net probably enhances
efficiency by lowering the (shadow) cost of funds, especially in the MENA
region where the level of access to banking is particularly low.

110 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
4.4 Conclusions
The results of this section clearly show that the banks in countries with a
sound regulatory structure are significantly more efficient. In particular,
the presence of deposit insurance schemes, disclosure standards that
facilitate private monitoring and the availability of and access to credit
information all enhance the cost efficiencies of banks. The stringency of
capital requirements also has a positive impact on bank efficiency, albeit to
a less extent. In turn, according to our findings, there is a case for allowing
banks to engage in a wider scope of activities and dismantling entry
obstacles. Supervisory independence seems to have no impact on cost
efficiency, most likely due to the offsetting impact of increased risks arising
from concentrated political power.
In short, our results support mainly the third pillar of Basel II with
weaker support for the capital requirements. The rapid deployment of
private credit bureaus, possibly modelled after the regional best-practice as
evidenced by Morocco’s brand new system, is also important in enhancing
efficiency. However, none of these factors should be treated in a vacuum.
Institutional quality, measured here by an aggregation of a number of
political and governance-related factors, is a substantially important factor.
Lastly, macroeconomic stability is also an important contributor to the
efficiencies of banks.
It should be highlighted that the results of this section have assessed
the importance of regulatory and supervisory practices for achieving bank
efficiency. Other issues should also be considered for making a broad
assessment of the suitability and adequacy of certain rules and standards.
For example, while certain regulatory conditions may improve banks’ cost
efficiencies, they may undermine profits (e.g. systemic stability).

5. IMPACT OF BANK REGULATIONS ON
GROWTH
T
he key justification for introducing financial regulations is based on
the idea that financial markets are imperfect and that regulations can
effectively correct these shortcomings. The various areas of
regulations considered in this study have all sprung into existence
due to these considerations. Since customers often have asymmetric
information regarding the operations of the banks, licensing and disclosure
requirements are put forward to restrict the possibility of improper
activities while providing the investors with adequate information. Capital
requirements are an attempt to contain the risk-taking incentives of the
owners of banks. The powers granted to the supervisors ensure that they
have access to adequate information on the financial intermediaries and
can act in a timely and efficient manner when troubles arise. Deposit
insurance schemes are put forward to mitigate the likelihood that
imperfectly informed depositors lead to a bank run. Credit information
availability is crucial to overcome credit rationing, which arises when the
financial intermediaries have limited information on borrowers.
If the regulations and supervisory practices serve to respond to
market imperfections in practice, they should have a clear pro-growth
impact. A better functioning financial market that properly treats the
information asymmetries that exist between the banks, their clients and the
supervisors should indeed help allocate financial resources more
efficiently. However, the impact of regulations may be more insidious if the
authorities choose to use their powers for their own good and not for the
common welfare. Under this so-called ‘private interest view’, politicians
may attempt to orient the industry to lend to their politically connected
clients and banks may capture the regulators to act in their own interests.
| 111

112 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
In short, the political imperfections may pose a greater risk to growth than
the market imperfections. 82
This chapter turns to a broader investigation of the economic benefits
of regulatory and supervisory practices. The main question is whether
banking regulations and practices have an impact on growth. Several
channels through which the relationship may operate are considered,
including the impact of regulations on cost efficiency, issuance of credit to
the private sector and capital market activity. The empirical analysis also
controls for the presence of other intermediate channels that are not
accounted for.
The results show that regulations impact economic growth through
their impact on bank efficiency and financial development. The role of
government in the banking sector has a clear negative impact on growth,
even beyond its impact on the identified intermediate variables. Thus,
governments that are heavily present in the banking sector also engage in
other activities that are less favourable to growth. Moreover, the impact of
disclosure requirements, scope restrictions and capital requirements are
mostly indirect, operating through the financial development variables.
5.1 Literature review
One of the key links between financial regulations and growth is the
presence of entry obstacles as a key impediment to a competitive market.
There are several conduits through which financial regulations may exert
an impact on economic growth. A well-functioning regulatory framework
can reinforce financial development and, in doing so, facilitates the flow of
funding to the real sectors. The predominant view in economic literature is
that a developed financial system can generate significant benefits for the
economy. Although the idea that the development of the financial services
sector is essential for economic development goes at least as far back as
Schumpeter (1934), it was King & Levine (1993) who first empirically
demonstrated a strong and robust relationship. Since then, several studies
have confirmed that financial development enhances growth through the
availability of external funds, higher employment, firm creation, etc. 83
82 See Barth el al. (2006, pp. 21-46 and 178-280) for more details on the public- and
private-interest views to financial regulations.
83 For a more complete survey of the so-called ‘finance-growth’ literature, see
Levine (1997; 2004) and Demirgüç-Kunt and Levine (2008).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 113
Economic theory identifies two main channels through which the
positive impact of financial development on growth operates. On the one
hand, a sound financial system increases the availability of resources for
investment by mobilising idle savings, facilitating transactions and
attracting foreign investments. On the other hand, such a system can
improve the allocation of funding by enhancing risk management,
transparency and corporate governance practices; reinforcing property and
creditor rights; and so forth. A well-functioning financial system is
particularly important for the development of the private sector,
particularly small and medium-sized enterprises, which represent a
significant proportion of economic activity but lack the internal sources to
grow. 84
Despite its wide recognition, several studies have challenged the
validity of the so-called ‘finance-growth’ view. Most of these doubts rest on
the direction of causality. 85 In particular, using a panel of less developed
countries, Demetriades & Hussein (1996) find evidence of bidirectionality—and
in some cases inverse correlation—for a panel of 16
developing countries. 86 Allowing for a non-linear relationship, Deidda &
Fattouh (2002) fail to confirm the results of King and Levine (1993) for less
developed countries included in their dataset. Similarly, Rioja & Valev
(2004) verify that the relationship depends on the level of economic
development, with little or uncertain impact on low or high extremes of the
income levels. Using a dataset of 11 MENA countries over the 1979-2003
period, Ben Naceur & Ghazouani (2007) find that the development of
banking and the stock market has no – or even a negative – impact on
growth in the MENA region.
One explanation of these contrasting results is that some factors that
are unaccounted for in the empirical analysis may explain why a financial
system functions well and economic growth occurs (possibly
84 Examining a number of national surveys, Fadil (2000) finds that a general lack of
access to credit markets is one of the principal constraints faced by SMEs to grow
in line with their cash-flow in the MENA region.
85 For a review of criticism of studies linking finance development to growth, see
Wachtel (2001; 2003) and references therein. See also Arestis & Demetriades (1997)
for reasons on why cross-country empirical studies may suffer from serious
methodological problems.
86 The only MED-11 country included in the sample of Demetriades & Hussein
(1996) is Turkey.

114 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
simultaneously). The omission of these variables may then lead to an
incorrect assessment of the direction and strength of the causal
relationship. 87 In response to these criticisms, the literature has turned on
these deeper structural conditions. Indeed, the emerging academic
consensus is that financial development could be beneficial as long as
certain conditions are present to ensure that the system develops
adequately to serve the financial needs of the citizens and the private
sector.
Development of financial regulations can also impact economic
growth through their effect on efficiency and competitive conditions in the
financial sector. The previous section has also given some evidence of the
impact of financial regulations on cost efficiencies of the banks in the
Mediterranean region. Although the literature on the impact of regulations
on bank efficiency is currently at its infancy, several studies have reached
similar conclusions. In particular, Barth et al. (2006) uses the results of their
own regulatory and supervisory surveys (BRSS) to provide partial support
for the positive impact of disclosure requirements on net interest margins
and cost efficiency. Pasiouras (2008) and Pasiouras et al. (2009) also use
same data source to confirm the results of Barth et al. (2006) while showing
that certain bank- and market-specific factors also matter. 88
A number of studies have also noted how regulatory conditions may
impact entry and more broadly the competitive conditions in financial
markets. Focusing on entry into banking markets, Cetorelli & Strahan
(2006) find that state-level restrictions on bank entry reduces the share of
smaller enterprises, effectively reducing the growth potential of the state.
Larger firms are less affected by entry obstacles, as they can use alternative
funding sources and have an easier access to capital markets. Restricted
entry tends to support the market power of the incumbent firms, which
could reduce the credit available to the economy as a whole and thereby
have a negative impact on growth, (Cetorelli & Gambera, 2001). 89 Using a
87 The idea that unaccounted for factors and relationships between non-structural
variables may introduce biases in the empirical assessment of policy impacts goes
back to Lucas (1976).
88 See section 4.1 for a more detailed treatment of the literature.
89 Although greater concentration in the banking sector may reduce the overall
availability of credit, Cetorelli & Gambera (2001) find that it may enhance funding
for firms that specialise in research and development, are highly dependent on
external finance and develop long-lasting relationships with their creditors. These

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 115
large sample of banks from the Middle East and North African countries,
Turk-Ariss (2009) shows that the degree of competition, measured by the
so-called ‘H-statistic’ first developed by Panzar & Rosse (1987), is positively
correlated with foreign bank entry but is a decreasing function of activity
restrictions. 90
5.2 Methodology and data
The basic regression used in this section takes the following functional
form:
where Growth is the real per capita growth for country i at year t. The
variables F, C, and R represent financial development, macroeconomic and
regulatory factors. In order to capture non-linear relationships, the natural
logarithms forms were used for most economic and financial variables.
The use of financial development variables as explanatory variables
may pose a bias in our estimations, since economic growth and financial
development may be determined simultaneously. To control for these
potential problems, an instrumental variables approach has been used in
this section. As in La Porta et al. (1997), legal origin (i.e. French, English or
mixed) is assumed to shape financial development. The use of legal origins
as an instrument for financial development has been a popular tool since
these institutional conditions can be safely treated as a purely exogenous
(i.e. unchanging) determinant of economic growth. Moreover, several
studies have shown evidence that legal origins influence financial
development through their impact on the treatment of shareholders, rights
of creditors, effectiveness of contract enforcement and the use of
international accounting standards. More specifically, La Porta et al. (1998)
show that French civil law countries are relatively low performing in terms
of shareholder and creditor rights, with less comprehensive accounting
issues are less likely to be applicable in the Middle East and North African
perspective since reliance on external financing is relatively low.
90 Using 2000-06 figures, Turk-Ariss (2009) finds that market conditions in Algeria,
Morocco and Tunisia can best be categorised as a monopoly. The banks in other
MENA countries included in the study (Egyptian banks were not covered in the
sample) operate under monopolistic competition.

116 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
standards. 91 Two dummy variables, English or mixed legal origin, are
included to account for different types of systems. Additionally, a dummy
variable for Muslim countries is also included.
Assuming that legal origin indicators serve as appropriate
instruments of financial development variables (F) is equivalent to a set of
orthogonality conditions for the instrumental variables on the instrument
variables (Z) and the error term, . Two-staged Generalised
Method of Moments (GMM) techniques are used to estimate the models
with the relevant orthogonality conditions.
Since the number of moment conditions may exceed the number of
coefficients to be estimated, tests of over-identifying restrictions are carried
out. These tests determine whether or not the instrumental variables are
associated with growth beyond their ability to explain any variation in
financial sector development. More specifically, the Hansen-Sargan test (‘Jtest’)
has a null hypothesis of correct model specification, which has an
asymptotic χ 2 distribution with degrees of freedom of the number of overidentifying
restrictions (Hansen, 1982). Failure to reject the test supports the
validity of the model.
A second set of tests is also carried out to check the weakness or
strength of the instruments. The so-called ‘Cragg-Donald test’ is simply an
F-statistic on the hypothesis that the instruments do not enter the first stage
regression of the two-stage estimations. A failure to reject the null
hypothesis calls into question the validity of the instrumental variable
estimates and hypothesis tests. The critical values of the test are given in
Stock & Yogo (2001). As a simple rule of thumb, specifications with a
Cragg-Donald F-value that exceeds 9.08 will be considered to be
appropriately defined. 92
Table 5.1 Data sources and descriptive statistics
Variable name Source
Obs. Mean
St.
dev.
Min Max
91 These results are by and large confirmed in our sample. The three countries with
elements of common British law (Cyprus, Israel and Malta) score very high on
creditors’ rights and the certified audit requirement for banks.
92 For our purposes, the Cragg-Donald threshold of 9.08 (with three instruments
and a single endogenous variable) corresponds to a maximum bias of 10% at the
5% significance level. For more information, see Stock & Yogo (2001, Table 1).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 117
Real GDP per capita growth World Dev. Ind. (WDI) 142 2.424 2.081 -3.607 10.577
Initial GDP per capita (log) WDI 154 8.603 1.064 7.028 9.776
Trade openness (log) WDI 153 -0.343 0.381 -0.958 0.700
Inflation (log) WDI 154 0.037 0.031 -0.004 0.261
Lack of corruption (log) PRS Group 154 1.116 0.343 0.405 1.609
Bank efficiency (log) Own calc. and Bankscope 154 0.519 0.091 0.321 0.692
Private credit (log) WDI 153 -0.389 0.829 -3.242 0.945
Stock market turnover (log) WDI 140 -1.215 1.165 -4.143 0.947
Scope restrictions Barth et al. surveys (BRSS) 137 7.526 1.595 5.000 10.000
Government ownership Barth et al. surveys (BRSS) 78 0.275 0.303 0.000 0.958
Cap. req. stringency Barth et al. surveys (BRSS) 137 4.869 2.141 1.000 9.000
Supervisory independence Barth et al. surveys (BRSS) 154 1.760 0.893 0.000 3.000
Deposit insurance Barth et al. surveys (BRSS) 137 0.766 0.425 0.000 1.000
Private monitoring Barth et al. surveys (BRSS) 137 8.212 0.958 6.000 10.000
Credit information Doing Business surveys 140 8.000 2.921 4.000 14.000
Notes: Bankscope database is compiled and distributed by Bureau van Dijk; World
Development Indicators (WDI) and Doing Business surveys are both distributed by the
World Bank.
The regulatory variables (R) are only available for the years 2000,
2003 and 2007. Nevertheless, changes in regulations are relatively slow
over time, as the results in section 2 amply demonstrated. It is therefore
reasonable to assume that the regulatory factors remain constant in the
periods prior to the observed outcomes, i.e. in 1995-2000, 2001-03 and 2004-
07.
The sources and descriptive statistics for the data used in this section
are summarised in Table 5.1. A number of country-specific time variant
variables are used to control for macroeconomic factors. Real initial GDP
per capita is included to account for fast growth in poorer countries. The
expected sign for the coefficient is negative, implying a significant catch-up
effect. Openness to trade, which is calculated as the imports and exports
divided by GDP, accounts for the positive spillovers from an open current
account. Inflation rate is included to account for the impact of economic
instability or inflationary policies on growth. Lastly, lack of corruption
assesses the level of corruption within the political system and the
bureaucracy.

Table 5.2 Impact of regulations on growth controlling for efficiency
I II III IV V VI VII VIII
Bank efficiency 6.118* 0.700 0.038 3.623 6.584 2.824 2.749 1.880
Initial GDP per
capita (real)
(2.956) (3.288) (3.707) (3.782) (4.004) (3.316) (3.080) (3.150)
-0.920*** -0.977*** -1.328*** -0.928*** -0.878*** -0.876*** -0.974*** -0.907***
(0.269) (0.271) (0.313) (0.319) (0.275) (0.279) (0.285) (0.305)
Openness to trade 0.659 0.089 -0.172 0.268 0.661 0.104 0.148 0.993
(0.490) (0.472) (0.828) (0.500) (0.461) (0.456) (0.472) (0.621)
Inflation 3.883 6.885 6.599 5.979 4.119 5.696 7.364 2.991
(6.866) (7.168) (8.575) (7.678) (6.596) (7.668) (7.244) (6.313)
Lack of corruption 1.130 1.626** 0.169 1.635** 0.961 1.490** 1.412** 1.118*
(0.722) (0.637) (0.946) (0.702) (0.853) (0.654) (0.673) (0.678)
Scope restrictions -0.242* .. .. .. .. .. ..
Governmentownership
Cap. req.
stringency
(0.134)
.. .. -0.023* .. .. .. .. ..
(0.013)
.. .. .. -0.160 .. .. .. ..
(0.097)
Supervisory .. .. .. .. 0.068 .. .. ..
| 118

independence
CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 119
(0.223)
Deposit insurance .. .. .. .. .. -0.358 .. ..
Private
monitoring
(0.427)
.. .. .. .. .. .. 0.301 ..
(0.189)
Credit information .. .. .. .. .. .. .. 0.037
(0.092)
Constant 5.922** 10.219*** 13.522*** 7.290*** 5.362* 6.911*** 5.096* 7.968***
(2.688) (2.384) (3.306) (2.531) (2.872) (2.392) (2.750) (2.525)
Observations 142 127 77 127 142 127 127 130
F-test (secondstage)
4.068 4.907 4.555 2.303 3.599 2.712 2.928 4.253
… p-value 0.002 0.000 0.001 0.039 0.002 0.017 0.011 0.001
Hansen J-test for
overidentification
2.057 2.955 2.178 2.824 1.961 0.704 0.448 3.075
… p-value 0.358 0.228 0.337 0.244 0.375 0.703 0.799 0.215
Cragg-Donald test
for weak id.(>9.08)
10.01 13.62 13.71 13.56 9.92 15.70 19.91 16.34

120 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
Table 5.3 Impact of regulations on growth controlling for private credit
I II III IV V VI VII VIII
Private credit -0.631 -0.839 -1.125 -0.953 -3.856 -0.433 -1.034 -0.196
(0.984) (0.780) (1.194) (0.780) (3.105) (0.761) (0.957) (0.475)
Initial GDP per capita (real) -0.769** -0.753** -1.299*** -0.656* 0.402 -0.780** -0.796** -0.923***
(0.382) (0.335) (0.286) (0.336) (1.103) (0.307) (0.318) (0.300)
Openness to trade 0.671 0.431 -0.083 0.475 1.518 0.139 0.448 0.966
(0.679) (0.568) (0.667) (0.559) (1.151) (0.517) (0.599) (0.664)
Inflation -0.181 5.856 13.660 4.545 -15.272 3.945 5.817 0.952
(7.568) (7.393) (9.036) (7.867) (19.562) (7.503) (7.115) (6.685)
Lack of corruption 1.947** 2.100*** 0.977 2.292*** 1.922* 1.912** 2.180*** 1.295*
(0.867) (0.754) (1.175) (0.849) (1.027) (0.841) (0.828) (0.690)
Scope restrictions .. -0.200 .. .. .. .. .. ..
(0.136)
Government-ownership .. .. -0.042* .. .. .. .. ..
(0.024)
Cap. req. stringency .. .. .. -0.102 .. .. .. ..
(0.072)
Supervisory independence .. .. .. .. 1.498 .. .. ..
(1.254)

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 121
Deposit insurance .. .. .. .. .. -0.306 .. ..
(0.418)
Private monitoring .. .. .. .. .. .. 0.514** ..
(0.250)
Credit information .. .. .. .. .. .. .. 0.074
(0.109)
Constant 6.897* 7.739** 12.530*** 5.691* -6.144 7.024*** 2.297 8.609***
(3.652) (3.220) (2.822) (3.024) (12.352) (2.639) (4.276) (2.030)
Observations 142 127 77 127 142 127 127 130
F-test (second-stage) 3.986 5.087 5.713 2.318 2.470 2.747 3.055 4.284
… p-value 0.002 0.000 0.000 0.038 0.027 0.016 0.008 0.001
Hansen J-test for
overidentification
4.367 1.975 2.053 2.400 0.659 1.151 0.107 3.264
… p-value 0.113 0.372 0.358 0.301 0.719 0.562 0.948 0.196
Cragg-Donald test for weak
id.(> 9.08)
3.457 8.163 3.556 6.148 4.475 7.127 4.779 15.80

122 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
Table 5.4 Impact of regulations on growth controlling for stock market turnover
I II III IV V VI VII VIII
Stock turnover 0.784** 0.393 0.312 0.577 0.779** 0.393 0.394 0.657
(0.363) (0.400) (0.356) (0.415) (0.363) (0.398) (0.385) (0.463)
Initial GDP per capita (real) -1.306*** -1.179*** -1.712*** -1.225*** -1.297*** -1.103*** -1.155*** -0.971***
(0.289) (0.334) (0.376) (0.398) (0.296) (0.346) (0.332) (0.300)
Openness to trade 1.715** 0.854 0.360 1.194 1.868** 0.694 0.774 1.610*
(0.805) (0.902) (1.227) (0.938) (0.774) (0.862) (0.885) (0.840)
Inflation 4.864 5.737 12.913 5.510 6.756 4.806 7.024 9.622
(6.482) (6.762) (9.510) (7.437) (5.957) (6.830) (6.928) (7.145)
Lack of corruption 1.156* 1.320* 0.302 1.933*** 1.303* 1.731*** 1.554** 1.220*
(0.696) (0.730) (1.044) (0.735) (0.787) (0.638) (0.686) (0.672)
Scope restrictions .. -0.225 .. .. .. .. .. ..
(0.153)
Government-ownership .. .. -0.043*** .. .. .. .. ..
(0.015)
Cap. req. stringency .. .. .. -0.149* .. .. .. ..
(0.081)
Supervisory independence .. .. .. .. -0.186 .. .. ..
(0.218)

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 123
Deposit insurance .. .. .. .. .. -0.438 .. ..
(0.447)
Private monitoring .. .. .. .. .. .. 0.274 ..
(0.205)
Credit information .. .. .. .. .. .. .. -0.130
(0.126)
Constant 13.785*** 13.357*** 17.426*** 12.341*** 13.844*** 10.841*** 8.839** 11.496***
(2.685) (2.971) (4.388) (4.128) (2.771) (3.534) (4.050) (2.610)
Observations 129 119 69 119 129 119 119 117
F-test (second-stage) 6.926 5.521 6.895 3.516 5.686 3.675 3.489 5.350
… p-value 0.000 0.000 0.000 0.003 0.000 0.002 0.003 0.000
Hansen J-test for
overidentification
0.287 2.377 3.290 2.647 0.536 0.730 0.352 2.011
… p-value 0.866 0.305 0.193 0.266 0.765 0.694 0.839 0.366
Cragg-Donald test for weak
id.(> 9.08)
11.07 9.827 15.01 8.701 11.97 9.172 11.76 15.28

CONVERGENCE OF BANKING SECTOR REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 124
Three financial variables are used to assess the impact on growth.
First, bank efficiency, which is the country average for the metatechnology
ratio scores for the banks (developed in section 2). Second,
private credit measures the share of private credits to the GDP. Lastly, the
stock turnover measures the ratio of stocks traded divided by the average
market capitalisation for the period.
In addition to these variables, the regulatory variables already
revised in section 4 are included, including scope restrictions, governmentowned
banks, capital requirements stringency, supervisor independence,
deposit insurance, private monitoring and credit information. Entry
obstacles were not included in the tests as government-ownership due to a
lack of observations for foreign denials. Instead, the market share of stateowned
banks is used as an indicator of the entry conditions, including
foreign denials and the number of licensing requirements. 93
5.3 Results
The results of the regressions are summarised in Table 5.2. The Hansen Jtests
fail to reject the null hypotheses for the 24 specifications, implying
that the instruments are properly used and are not correlated with the
residuals of the (second-stage) regressions. An additional statistic, the
Cragg-Donald test for weak identification, is also included. These results
point at potential problems due to weak instruments in Table 5.3, which
endogenously account for private credit except for the last column (VIII),
which controls for the level of credit information available.
The results show that some of the country-specific variables do not
matter while others have a significant impact on growth. In particular, our
results reveal that inflation is not a significant determinant of growth in
any of the specifications, despite a consistently positive coefficient estimate.
Openness to trade has a weak positive impact on growth when stock
market turnover is considered in Table 5.4, i.e. columns I and VIII. In turn,
initial GDP per capita also has a persistently and significantly negative
impact on growth, which implies that poorer countries tend to grow more
quickly than richer ones, as in Barro (1991).
93 For more details on entry obstacles, see secion 2.3. The pairwise correlation
coefficients between the market share of state-owned banks on the one hand and
the licensing requirement and foreign denial scores on the other are 0.270 and
0.435, respectively.

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 125
There is broad evidence that lack of corruption has a relatively
consistent and positive impact on growth. This is in line with the findings
in the growth literature. 94 Corruption could lead to a number of
inefficiencies, such as rent-seeking and avoidance behaviour, which may
lead to substantial deadweight losses (Rose-Ackerman, 1975; Shleifer &
Vishny, 1993). Mauro (1995) shows that corruption affects growth by
lowering returns from private investment.
The results show that the three financial development variables have a
relatively limited impact on growth. In particular, private credit (Table 5.3)
does not have a significant impact (although all of the coefficient estimates
are negative) when it is allowed to be the endogenously determined
financial development variable. In turn, both bank efficiency (Table 5.2)
and stock market turnover (
94 Several studies linking the impact of financial development to growth have used
black market premia to account for corruption-related factors (Beck et al., 2000;
Beck & Levine, 2004; Ben Naceur & Ghazouani, 2007).

126 | IMPACT OF BANK REGULATIONS ON EFFICIENCY
Table 5.4) have positive impacts on growth.
Among the regulatory factors, government ownership has a weak but
consistent impact, reducing growth in all three tables. More specifically,
countries in which the state-owned banks are predominant grow less
quickly. These findings are by and large supported in the literature. Barth
et al. (1999) find evidence that government ownership of banks is
associated with a low level of financial development, as measured by the
available of credit to private enterprises. La Porta et al. (2002) find that
government ownership of banks is associated with lower subsequent
financial development and growth in per capita income. Beck et al. (2004)
use firm-level data to show that public bank ownership tends to
exacerbates market power of the incumbents and thus constrains credit to
private enterprises. 95
Other regulatory factors have a less consistent impact. The strongest
impact is with private monitoring, which is positively associated with
capital income growth when private credit is considered but not in other
cases. Scope restrictions tend to have a negative impact on growth when
bank efficiency is considered but not when the two financial development
factors are considered. Lastly, capital requirement stringency has a weak
negative impact when the presence of stock market turnover is considered
as an endogenous variable.
These findings imply that the government’s role most likely serves as
a proxy for other activities that are detrimental to growth—possibly
unrelated to the financial markets. In turn, the impact of private
monitoring, scope restrictions and capital requirements are mostly indirect,
operating through the financial development variables, since their
independent effects become insignificant when some of the financial
development factors are controlled for.
95 Part of the literature argues that excessive government ownership can be
harmful because politicians use government-owned banks to further their own
political goals (Shleifer & Vishny, 1994; La Porta et al., 2002). Dinc (2005) provides
support to this ‘political view’ that public bank policies are often politicallyoriented
in finding that public banks increase their lending in election years.
Caprio & Peria (2000) show that state ownership of banking is associated with a
greater likelihood of crises. The latter finding should also be considered in light of
evidence that public banks in developing countries are often more stable than their
privately-owned counterparts (Ayadi et al., 2009).

CONVERGENCE OF BANK REGULATIONS IN THE SOUTHERN MEDITERRANEAN | 127
5.4 Conclusions
According to our specifications, financial regulations have a relatively
limited direct impact on growth. Among the seven regulatory areas
considered throughout the paper, only government ownership—a proxy
for entry obstacles and market conditions—appears to have a consistent
and significant negative impact on growth. Moreover, there is limited
evidence that financial development leads to economic development.
Although efficiency and stock market turnover appear to increase income
per capita growth, private credit appears to have little (and possibly
negative) impact.
These results show that regulatory factors operate mostly through the
financial variables. Based on the results reviewed in this section, the
regulatory factors considered in the paper have at best an indirect impact
on growth, working their way through financial development. Moreover,
lack of corruption has a clear impact on growth, which underlines its
importance as a precondition for growth.
Several technical shortcomings have to be noted at this stage. First,
due to the small sample size considered in the study, panel estimations
were not feasible. Second, the similarities between the countries considered
might have generated sampling biases, which imply that the results have to
be interpreted with care and should be adequately re-assessed before
applying to other regions. Third, the similarities between countries also
make the task of finding strong indicators more difficult as the crosscountry
variation is relatively limited. This is indeed one of the main
causes for the apparent weaknesses of the instruments for the share of
private credit in GDP. Lastly, the regulatory variables are assumed to
remain fixed over long periods; although this assumption is unlikely to
lead to substantial biases, it creates another source of homogeneity.

6 CONCLUSIONS
T
his study sheds light on the changing regulatory environments of
four south Mediterranean countries: Algeria, Egypt, Morocco
and Tunisia. Over the past two decades, all four countries have
engaged in financial sector reforms, with varying degrees of
depth, engagement and success. Morocco has achieved the most
advanced financial system as compared to the three others, eliminating
interest rate subsidies and controls; reinforcing the responsibilities and
roles of the supervisor; improving the risk-management practices along
with the state-of-the-art; successfully implementing of a deposit
insurance scheme; and introducing a credit information system that
may well serve as a best-practice for other developing financial
systems.
The other South-MED countries examined in this study have been
less successful in implementing key reforms. The banks in all of the
three countries have relatively poor asset qualities, as evidenced by
high rates of non-performing loans (NPLs). The policies put in place to
respond to low asset quality have either led to limited improvement, a
decline of credit availability or both. The privatisation efforts have been
only partly successful and at times have not led to any change in the
market conditions and financial development. In Algeria, the publiclyowned
banks continue to dominate the banking sector, accounting for
over 90% of total assets. In Egypt, although privatisation efforts have
been partly successful, public loans and debt represent a substantial
proportion of the portfolios of banks, which hampers financial
development and growth opportunities. In Tunisia, a majority of the
top three banks remain owned by the state.
The comparisons among the EU-MED countries reveal particular
shortcomings. Despite some recent improvements, entry obstacles
continue to be widespread in all of the South-MED countries, arising
from high rates of denied foreign applications and closely linked with a
| 128

dominant state ownership. Capital requirements are less stringent in
the Southern Mediterranean under examination, increasingly so due to
the disparities in the risk-sensitivity of the minimum capital
requirements. The existing deposit insurance schemes in place, i.e.
those in Algeria and Morocco, provide adverse incentives and may
increase moral hazard risks. In Egypt and Tunisia, the implicit
government guarantees may also aggravate the moral hazard problem.
Although private monitoring and disclosure requirements appear in
line with the EU-MED standards, accounting practices are increasingly
poor in the South-MED. Lastly, despite recent improvements,
especially in Morocco and Egypt, credit information availability is
relatively low within the region.
Turning to the cost efficiency analysis, results indicate an overall
improvement in efficiency levels for the EU-MED and South-MED in
the later stages of the analysis, from 2005 onwards (with the exception
of Egypt). For the South-MED, this improvement is particularly
remarkable for Moroccan and Algerian banks, but for different reasons.
The overall mean efficiency in the region is improving, once more
driven by improvements in the best practice. EU-MED banks, in
particular the Spanish banks, dominate the region, with average
efficiency scores of 80.4% against the region's average of 63.5%. Spanish
banks also exhibit the highest meta-technology ratios and the ratios
increase over time. This indicates that Spanish banks consistently
improved their performance, and their banking technology became best
practice. Nonetheless, during this period of analysis, the average metatechnology
ratio is increasing, which indicates an ability of banks in all
countries to appropriate the best available technology. These results are
supported by the estimation of !-convergence. The ! coefficient is
always negative and statistically significant, thus indicating that
convergence in efficiency scores has occurred across countries in the
MED-11 area. Furthermore, results for the "-convergence suggest an
increase in the speed of convergence as the " coefficient is always
negative and statistically significant. This indicates that, whereas the
technological gap is still wide, the gap is narrowing at a faster speed.
When examining the impact of the regulatory and supervisory
practices on cost efficiency of banks, the results clearly show that a
sound regulatory structure is a forceful contributor to an efficient
system. The case of Morocco is revealing in this respect. In particular,
deposit insurance schemes, adequate disclosure requirements and
credit information availability seem to improve the efficiencies of
129

130 | CONCLUSIONS
banks. A broader definition of the banking market by imposing fewer
scope restrictions and removing entry obstacles also improves
efficiency, albeit less significantly so than the previous factors. The
rapid deployment of private credit bureaus, possibly modelled after the
regional best-practice as evidenced by Morocco’s brand new system, is
also important in enhancing efficiency.
Lastly, the pro-growth impact of regulatory adequacy appears to
operate mainly though its impact on financial development. The study
shows that government ownership in banking is detrimental to growth,
even outside the scope of financial development and other financerelated
variables. Moreover, more restrictive disclosure and capital
requirements as well as less limited scope restrictions have pro-growth
impacts by enhancing financial development.
It is important to note that the regulatory practices and adequacy
factors should not be treated in a vacuum. Institutional quality,
measured in the study by a variety of political and governance-related
factors, is a substantially important factor in all of the regressions. The
control of corruption and the presence of democratic institutions are
also important factors, which need to be considered alongside the
regulatory conditions.
To sum up, the study highlights some of the key shortcomings of
the banking regulations of the South-MED countries. It appears that
some of the newer standards, such as the Basel II capital requirements,
have been conceived with developed nations in mind and may not be
appropriate, due to a variety of deficiencies in information-sharing and
institutional and disclosure mechanisms. A key aim of the upcoming
reforms should be to look for ways to reduce the role of government in
the banking sector while ensuring that the regulatory framework and
the relevant institutional development adequately respond to the
market imperfections.
130

REFERENCES
Allain, L. and N. Oulidi (2009), “Credit Market in Morocco: A
Disequilibrium Approach“, IMF Working Paper No. 09/53,
International Monetary Fund, Washington, D.C.
Andrianova, S., P. Demetriades and A.K. Shortland (2010), “Government
Ownership of Banks, Institutions and Economic Growth“,
Discussion Papers in Economics, No. 11/01, Department of
Economics, University of Leicester.
Arellano, M. and O. Bover (1995), “Another look at the instrumental
variables estimation of error components models”, Journal of
Econometrics, Vol. 68.
Arestis, P. and P.O. Demetriades (1997), "Financial Development and
Economic Growth: Assessing the Evidence", Economic Journal, Vol.
107, No. 442.
Ayadi, R., R.H. Schmidt, S. Carbo Valverde, E. Arbak and F. Rodriguez
Fernandez (2009), Investigating Diversity in the Banking Sector in
Europe: The Performance and Role of Savings Banks, Centre for
European Policy Studies (CEPS), Brussels.
Baele, L.M., A. Ferrando, P. Hordahl, E. Krylova and C. Monnet (2004),
“Measuring financial integration in the euro area”, ECB Occasional
Paper No. 14, European Central Bank, Frankfurt.
Banerjee, A.V. and A.F. Newman (1993), “Occupational Choice and the
Process of Development“, Journal of Political Economy, Vol. 101, No. 2.
Banker, R.D. and R. Natarajan (2008), “Evaluating contextual variables
affecting productivity using Data Envelopment Analysis”, Operations
Research, Vol. 56, No. 1.
Banker, R.D., A. Charnes and W.W. Cooper (1984), “Some models for the
estimating technical and scale inefficiency in Data Envelopment
Analysis”, Management Science, Vol. 30.
Barro, R.J. (1991), “Economic Growth in a Cross Section of Countries“,
Quarterly Journal of Economics, Vol. 106, No. 2.
Barro, R.J. and X.X. Sala-i-Martin (1991), “Convergence across states and
regions”, Brookings Papers on Economic Activity, No. 1, Brookings
Institution, Washington, D.C.
Barro, R.J. and X.X. Sala-i-Martin (1992), “Convergence”, Journal of Political
Economy, Vol. 100.
|131

132 | ANNEX
Barro, R.J. and X.X. Sala-i-Martin (1995), Economic growth, New York:
McGraw Hill.
Barth, J.R., G. Caprio and R. Levine (2001), “The Regulation and
Supervision of Banks Around the World: A New Database“, Policy
Research Working Paper No. 2588, World Bank, Washington, D.C.
Barth, J.R., G. Caprio and R. Levine (2004), “Bank regulation and
supervision: What works best? “, Journal of Financial Intermediation,
Vol. 13, No. 2.
Barth, J.R., G. Caprio and R. Levine (2006), Rethinking Bank Supervision and
Regulation: Till angels govern, Cambridge: Cambridge University
Press.
Barth, J.R., G. Caprio and R. Levine (2008), “Bank regulations are changing:
But for better or worse? “, Policy Research Working Paper No. 4646,
World Bank, Washington, D.C.
Barth, J.R., G. Caprio, Jr. and R. Levine (1999), “Banking Systems around
the Globe: Do Regulation and Ownership Affect Performance and
Stability? “, Policy Research Working Paper No. 2325, World Bank,
Washington, D.C.
Battese, G.E. and T.J. Coelli (1995), “A model for technical inefficiency
effects in a stochastic frontier production function for panel data”,
Empirical Economics, Vol. 20.
Battese, G.E., D.S. Prasada Rao and C.J. O’Donnell (2004), “A Meta-frontier
Production Function for Estimation of Technical Efficiencies and
Technology Gaps for Firms Operating Under Different
Technologies”, Journal of Productivity Analysis, Vol. 21, No. 1.
BenNaceur, S., H. Ben-Khedhiri and B. Casu (2011), “What Drives the
Performance of Selected MENA Banks? A Meta-Frontier Analysis”,
IMF Working Paper No. 11 WP/11/34, International Monetary
Fund, Washington, D.C.
Berger, A.N. (2007), “International comparisons of banking efficiency”,
Financial Markets, Institutions and Instruments, Vol. 16, No. 3.
Berger, A.N. and D.D. Humphrey (1997), “Efficiency of Financial
Institutions: International Survey and Directions for Future
Research”, European Journal of Operational Research, Vol. 98.
Berger, A.N. and L.J. Mester (2003), “Explaining the dramatic changes in
performance of US banks: technological change, deregulation, and
dynamic changes in competition”, Journal of Financial Intermediation,
Vol. 12, No. 1.
132

Berger, A.N., R. DeYoung, H. Genay and G.H. Udell (2000), “Globalization
of Financial Institutions: Evidence from Cross-border Banking
Performance”, Brookings-Wharton Papers on Financial Services.
BCBS (Basel Committee on Banking Supervision) (2001), “Working Paper
on Pillar 3 – Market Discipline“, Bank for International Settlements
(BIS), Geneva, September.
Beck, T. and A. Demirgüç-Kunt (2009), “Financial Institutions and Markets
Across Countries and over Time: Data and Analysis“, Policy
Research Working Paper No. 4943, World Bank, Washington, D.C.
Beck, T., A. Demirguc-Kunt and V. Maksimovic (2004), “Bank Competition
and Access to Finance: International Evidence“, Journal of Money,
Credit, and Banking, Vol. 36, No. 3.
Beck, T. and R. Levine (2004), “Stock Markets, Banks, and Growth: Panel
Evidence“, Journal of Banking and Finance, Vol. 28, No. 3.
Beck, T., R. Levine and N. Loayza (2000), “Finance and the sources of
growth“, Journal of Financial Economics, Vol. 58, Nos. 1-2.
Ben Naceur, S., H. Ben-Khedhiri and B. Casu (2009), “What Drives the
Efficiency of Selected MENA Banks? A Meta-Frontier Analysis“,
Working Paper Series, No. 03/09, CASS Business School, City
University of London, Centre for Banking Research, London.
Ben Naceur, S. and S. Ghazouani (2007), “Stock Markets, Banks, and
Economic Growth: Empirical Evidence from the MENA Region“,
Research in International Business and Finance, Vol. 21, No. 2.
Berger, A.N. and T.H. Hannan (1998), “The Efficiency Cost of Market
Power in the Banking Industry: A Test of the 'Quiet Life' and Related
Hypotheses“, Review of Economics and Statistics, Vol. 80, No. 3.
Bhattacharya, S. and A.V. Thakor (1993), “Contemporary Banking Theory“,
Journal of Financial Intermediation, Vol. 3, No. 1.
Bikker, J.A. (2004), Competition and efficiency in a unified European banking
market, Cheltenham, UK: Edward Elgar.
Blundell, R. and S.R. Bond (1998), “Initial conditions and moment
restrictions in dynamic panel data models”, Journal of Econometrics,
Vol. 87.
Bos, J.W.B. and C.J.M. Kool (2006), “Bank Efficiency: The Role of Bank
Strategy and Local Market Conditions“, Journal of Banking and
Finance, Vol. 30, No. 7.
Bos, J.W.B. and H. Schmiedel (2007), “Is there a single frontier in a single
European banking market?”, Journal of Banking and Finance, Vol. 31.
133

134 | ANNEX
Bos, J.W.B., F. Heid, M. Koetter, J.W. Kolari and C.J.M. Kool (2005),
“Inefficient or Just Different? Effects of Heterogeneity on Bank
Efficiency Scores”, Discussion Papers Series 2, Banking and Finance
Studies, Vol. 15, Deutsche Bank, Research Center.
Boyd, J.H., C. Chang and B.D. Smith (1998), "Moral Hazard under
Commercial and Universal Banking", Journal of Money, Credit, and
Banking, Vol. 30, No. 3.
Brown, C.O. and I.S. Dinc (2005), “The Politics of Bank Failures: Evidence
from Emerging Markets“, Quarterly Journal of Economics, Vol. 120,
No. 4.
Brown, M., T. Jappelli and M. Pagano (2009), “Information Sharing and
Credit: Firm-Level Evidence from Transition Countries“, Journal of
Financial Intermediation, Vol. 18, No. 2.
Caprio, G. and P. Honohan (2004), “Can the Unsophisticated Market
Provide Discipline? “, Policy Research Working Paper No. 3364,
World Bank, Washington, D.C.
Caprio, G. and M.S.M. Peria (2000), “Avoiding Disaster: Policies to Reduce
the Risk of Banking Crises“, ECES Working Paper No. 47, Egyptian
Center for Economic Studies (ECES), Cairo.
Casu, B. and C. Girardone (2010), “Integration and efficiency convergence
in EU banking markets”, OMEGA, The International Journal of
Management Science, Vol. 38, No. 5.
Cetorelli, N. and M. Gambera (2001), “Banking Market Structure, Financial
Dependence and Growth: International Evidence from Industry
Data“, Journal of Finance, Vol. 56, No. 2.
Cetorelli, N. and P.E. Strahan (2006), “Finance as a Barrier to Entry: Bank
Competition and Industry Structure in Local U.S. Markets“, Journal
of Finance, Vol. 61, No. 1.
Chaffai, M.E., M. Dietsch and A. Lozano-Vivas (2001), “Technological and
Environmental Differences in the European Banking Industries”,
Journal of Financial Services Research, Vol. 19.
Charnes, A. W.W. Cooper and E. Rhodes (1978), “Measuring efficiency of
decision making units”, European Journal of Operational Research, Vol.
2.
Chiuri, M.C., G. Ferri and G. Majnoni (2002), “The Macroeconomic Impact
of Bank Capital Requirements in Emerging Economies: Past
Evidence to Assess the Future“, Journal of Banking and Finance, Vol.
26, No. 5.
134

#ihák, M. and A.F. Tieman (2008), “Quality of Financial Sector Regulation
and Supervision Around the World“, IMF Working Paper No.
08/190, International Monetary Fund, Washington, D.C.
Claessens, S., A. Demirgüç-Kunt and H. Huizinga (2001), “How Does
Foreign Entry Affect Domestic Banking Markets? “, Journal of
Banking and Finance, Vol. 25, No. 5.
Claessens, S. and L. Laeven (2004), "What Drives Bank Competition? Some
International Evidence", Journal of Money, Credit, and Banking, Vol. 36,
No. 3.
Cole, S. (2009), "Fixing Market Failures or Fixing Elections? Agricultural
Credit in India", American Economic Journal: Applied Economics, Vol. 1,
No. 1.
Das, U.S., P. Iossifov, R. Podpiera and D. Rozhkov (2005), "Quality of
Financial Policies and Financial System Stress", IMF Working Paper
No. 05/173, International Monetary Fund, Washington, D.C.
Deidda, L. and B. Fattouh (2002), "Non-linearity between Finance and
Growth", Economics Letters, Vol. 74, No. 3.
Demetriades, P.O. and K.A. Hussein (1996), "Does financial development
cause economic growth? Time-series evidence from 16 countries",
Journal of Development Economics, Vol. 51, No. 2.
Demirgüç-Kunt, A. and E. Detragiache (2002), "Does Deposit Insurance
Increase Banking System Stability? An Empirical Investigation",
Journal of Monetary Economics, Vol. 49, No. 7.
Demirgüç-Kunt, A. and E. Detragiache (2010), "Basel Core Principles and
Bank Risk: Does Compliance Matter?", IMF Working Paper No.
10/81, International Monetary Fund, Washington, D.C.
Demirgüç-Kunt, A. and E.J. Kane (2002), "Deposit Insurance around the
Globe: Where Does It Work?", Journal of Economic Perspectives, Spring,
Vol. 16, No. 2.
Demirgüç-Kunt, A., B. Karacaovalı and L.A. Laeven (2005), "Deposit
Insurance Around the World Dataset", Policy Research Working
Paper No. 3628, World Bank, Washington, D.C.
Demirgüç-Kunt, A., L. Laeven and R. Levine (2004), "Regulations, Market
Structure, Institutions, and the Cost of Financial Intermediation",
Journal of Money, Credit, and Banking, Vol. 36, No. 3.
Demirgüç-Kunt, A. and R. Levine (2008), "Finance, Financial Sector
Policies, and Long-Run Growth", Policy Research Working Paper
No. 4469, World Bank, Washington, D.C.
135

136 | ANNEX
DeYoung, R. (1998), "Management Quality and X-Inefficiency in National
Banks", Journal of Financial Services Research, Vol. 13, No. 1.
Diamond, D.W. and P.H. Dybvig (1983), "Bank Runs, Deposit Insurance,
and Liquidity", Journal of Political Economy, Vol. 91, No. 3.
Dietsch, M. and A. Lozano-Vivas (2000), "How the Environment
Determines Banking Efficiency: A Comparison between French and
Spanish Industries", Journal of Banking and Finance, Vol. 24, No. 6.
Dinc, I.S. (2005), "Politicians and Banks: Political Influences on
Government-Owned Banks in Emerging Markets", Journal of
Financial Economics, Vol. 77, No. 2.
Djankov, S., C. McLiesh and A. Shleifer (2007), "Private Credit in 129
Countries", Journal of Financial Economics, Vol. 84, No. 2.
Emerging Markets Group (2006), "Strengthening Egypt’s Credit Reporting
System – Phase 2: Final Report", report prepared by Emerging
Markets Group, Ltd., prime contractor for the FIRST Initiative, under
Contract #49 C343, November.
European Commission (2010), "Impact assessment accompanying
document to the proposal on deposit guarantee schemes",
Commission Staff Working Document, SEC(2010) 834/2, Brussels.
Fadil, M.A. (2000), "A Survey of the Basic Features and Problems of the
Informal Small and Micro-Enterprises in the Arab Region", FEMISE
Research Programme.
Fries, S. and A. Taci (2005), "Cost Efficiency of Banks in Transition:
Evidence from 289 Banks in 15 Post-Communist Countries", Journal
of Banking and Finance, Special Issue, Vol. 29, No. 1.
Fung, M.K. (2006), “Scale economies, X-efficiency, and convergence of
productivity among bank holding companies”, Journal of Banking and
Finance, Vol. 30.
Galor, O. and J. Zeira (1993), "Income Distribution and Macroeconomics",
Review of Economic Studies, Vol. 60, No. 1.
Garcia-Marco, T. and M.D. Robles-Fernandez (2008), "Risk-Taking
Behaviour and Ownership in the Banking Industry: The Spanish
Evidence", Journal of Economics and Business, Vol. 60, No. 4.
Gerschenkron, A. (1962), Economic backwardness in historical perspective,
Cambridge, MA: Harvard University Press.
Goddard, J.A., P. Molyneux and J.O.S. Wilson (2001), European Banking:
Efficiency, Technology and Growth, London: John Wiley and Sons.
136

Goddard, J.A., P. Molyneux, J.O.S. Wilson and M. Tavakoli (2007),
“European banking: an overview”, Journal of Banking and Finance,
Vol. 31.
Grigorian, D.A. and V. Manole (2006), “Determinants of commercial bank
performance in transition: an application of data envelopment
analysis”, Comparative Economic Studies, Vol. 48.
Gropp, R. and J. Vesala (2004), "Deposit Insurance, Moral Hazard and
Market Monitoring", Review of Finance, Vol. 8, No. 4.
Hakenes, H. and I. Schnabel (2006), "The Threat of Capital Drain: A
Rationale for Public Banks?", Working Paper No. 11, Max Planck
Institute for Research on Collective.
Hansen, L.P. (1982), "Large Sample Properties of Generalized Method of
Moments Estimators", Econometrica, Vol. 50, No. 4.
Haselmann, R., K. Pistor and V. Vig (2010), "How Law Affects Lending",
Review of Financial Studies, Vol. 23, No. 2.
Hauner, D. (2008), "Credit to Government and Banking Sector
Performance", Journal of Banking and Finance, Vol. 32, No. 8.
Hauner, D. (2009), "Public Debt and Financial Development", Journal of
Development Economics, Vol. 88, No. 1.
Hicks, J.R. (1935), "Annual Survey of Economic Theory: The Theory of
Monopoly", Econometrica, Vol. 3, No. 1.
Houston, J.F., C. Lin, P. Lin and Y. Ma (2010), "Creditor Rights,
Information Sharing, and Bank Risk Taking", Journal of Financial
Economics, Vol. 96, No. 3.
Hughes, J.P., W.L. Lang, L.J. Mester, C. Moon and M.S. Pagano (2003), "Do
Bankers Sacrifice Value to Build Empires? Managerial Incentives,
Industry Consolidation, and Financial Performance", Journal of
Banking and Finance, Vol. 27, No. 3.
Hughes, J.P. and L.J. Mester (2010), “Efficiency in banking: theory, practise
and evidence”, in A. Berger, P. Molyneux and J.O.S. Wilson, The
Oxford Handbook of Banking, Oxford: Oxford University Press.
Huybens, E. and B.D. Smith (1999), "Inflation, Financial Markets and Long-
Run Real Activity", Journal of Monetary Economics, Vol. 43, No. 2.
IMF (2002), "Tunisia: Financial System Stability Assessment, including
Reports on the Observance of Standards and Codes on the following
topics: Monetary and Financial Policy Transparency, and Insurance
Regulation ", IMF Staff Country Reports, No. 02/119, International
Monetary Fund, Washington, D.C.
137

138 | ANNEX
–––––––– (2007), "Tunisia: Financial Sector Assessment Program Update -
Detailed Assessment of Compliance of the Basel Core Principles for
Effective Banking Supervision", IMF Staff Country Reports, No.
07/98, International Monetary Fund, Washington, D.C.
–––––––– (2008), "Morocco: Financial System Stability Assessment -
Update", IMF Staff Country Reports, No. 08/333, International
Monetary Fund, Washington, D.C.
–––––––– (2009), "Tunisia: 2009 Article IV Consultation—Staff Report", IMF
Country Reports, No. 09/329, International Monetary Fund,
Washington, D.C.
–––––––– (2010a), "Algeria: 2009 Article IV Consultation - Staff Report; and
Public Information Notice", IMF Country Report, No. 10/57,
International Monetary Fund, Washington, D.C.
–––––––– (2010b), "Arab Republic of Egypt: 2010 Article IV Consultation -
Staff Report Public Information Notice on the Executive Board
Discussion; and Statement by the Executive Director for the Arab
Republic of Egypt", IMF Country Report, No. 10/94, International
Monetary Fund, Washington, D.C.
–––––––– (2010c), "Morocco: 2009 Article IV Consultation - Staff Report;
Public Information Notice on the Executive Board Discussion; and
Statement by the Executive Director for Morocco", IMF Country
Report, No. 10/58, International Monetary Fund, Washington, D.C.
–––––––– (2010d), "Tunisia: 2010 Article IV Consultation—Staff Report;
Public Information Notice on the Executive Board Discussion; and
Statement by the Executive Director for Tunisia ", IMF Country
Report, No. 10/282, International Monetary Fund, Washington, D.C.,
September.
Isik, I. and M.K. Hassan (2003), “Financial deregulation and total factor
productivity change: an empirical study of Turkish commercial
banks”, Journal of Banking and Finance, Vol. 27.
Jappelli, T. and M. Pagano (2002), "Information Sharing, Lending and
Defaults: Cross-Country Evidence", Journal of Banking and Finance,
Vol. 26, No. 10.
Kane, E. (2000), "Designing financial safety nets to fit country
circumstances", Policy Research Working Paper No. 2453, World
Bank, Washington, D.C.
Kasman, A. and C. Yildirim (2006), "Cost and Profit Efficiencies in
Transition Banking: The Case of New EU Members", Applied
Economics, Vol. 38, No. 9.
138

Kaufmann, D., A. Kraay and M. Mastruzzi (2009), "Governance Matters
VIII: Governance Indicators for 1996-2008", World Bank Policy
Research, World Bank, Washington, D.C.
Khwaja, A.I. and A. Mian (2005), "Do Lenders Favor Politically Connected
Firms? Rent Provision in an Emerging Financial Market", Quarterly
Journal of Economics, Vol. 120, No. 4.
King, R.G. and R. Levine (1993), "Finance and growth: Schumpeter might
be right", Quarterly Journal of Economics, Vol. 108, No. 3.
La Porta, R., F. Lopez-de-Silane, A. Shleifer and R.W. Vishny (1997), "Legal
Determinants of External Finance", Journal of Finance, Vol. 52, No. 3.
La Porta, R., F. Lopez-de-Silanes and A. Shleifer (1998), "Law and Finance",
Journal of Political Economy, Vol. 106, No. 6.
La Porta, R., F. Lopez-de-Silanes and A. Shleifer (2002), "Government
Ownership of Banks", Journal of Finance, Vol. 57, No. 1, pp. 265-301.
Laeven, L. and R. Levine (2008), "Bank Governance, Regulation, and Risk
Taking", NBER Working Paper No. 14113, National Bureau of
Economic Research, Cambridge, MA.
Levine, R. (1997), "Financial Development and Economic Growth: Views
and Agenda", Journal of Economic Literature, Vol. 35, No. 2.
–––––––– (1998), "The Legal Environment, Banks, and Long-Run Economic
Growth", Journal of Money, Credit, and Banking, Vol. 30, No. 3.
–––––––– (2004), "Finance and Growth: Theory and Evidence", NBER
Working Paper No. 10766, National Bureau of Economic Research,
Cambridge, MA.
Liang, Z. (2006), "Financial Development and Income Distribution: A
System GMM Panel Analysis with Application to Urban China",
Journal of Economic Development, Vol. 31, No. 2.
Lucas, R.E., Jr. (1976), "Econometric Policy Evaluation: A Critique",
Carnegie-Rochester Conference Series on Public Policy, Vol. 1.
Madeddu, O. (2010), "The Status of Information Sharing and Credit
Reporting Infrastructure", MENA Financial Sector Flagship Report,
Improving Access to Finance while Maintaining Stability in the
Middle East and North Africa, World Bank, Washington, D.C.
Maudos, J. and J.F. de Guevara (2007), "The Cost of Market Power in
Banking: Social Welfare Loss vs. Cost Inefficiency", Journal of Banking
and Finance, Vol. 31, No. 7.
Mauro, P. (1995), "Corruption and Growth", Quarterly Journal of Economics,
Vol. 110, No. 3.
139

140 | ANNEX
Merton, R.C. (1977), "An analytic derivation of the cost of deposit
insurance and loan guarantees: An application of modern option
pricing theory", Journal of Banking & Finance, Vol. 1, No. 1.
MIX (2009), "Arab Microfinance Analysis and Benchmarking Report",
report from Microfinance Information Exchange (MIX), Sanabel and
the Consultative Group to Assist the Poor (CGAP).
Mukherjee, K., S.C. Ray and S.M. Miller (2001), “Productivity growth in
large us commercial banks: The initial post-deregulation
experience”, Journal of Banking and Finance, Vol. 25.
Murinde, V., J. Agung and A.W. Mullineux (2004), “Patterns of corporate
financing and financial system convergence in Europe”, Review of
International Economics, Vol. 12.
O’Donnel, C.J., D.S. Prasada Rao and G.E. Battese (2008), “Meta-frontier
Frameworks for the study of firm level efficiencies and technology
ratios”, Empirical Economics, Vol. 34.
Olson, D. and T.A. Zoubi (2010), "Efficiency and Bank Profitability in
MENA Countries", School of Business and Management, American
University of Sharjah, Sharjah, United Arab Emirates.
Pagano, M. and T. Jappelli (1993), "Information Sharing in Credit Markets",
Journal of Finance, Vol. 48, No. 5.
Panzar, J.C. and J.N. Rosse (1987), "Testing for 'Monopoly' Equilibrium",
Journal of Industrial Economics, Vol. 35, No. 4.
Parikh, A. and M. Shibata (2004), “Does trade liberalisation accelerate
convergence in per capita incomes in developing countries?”, Journal
of Asian Economics, Vol. 15.
Pasiouras, F. (2008), "International Evidence on the Impact of Regulations
and Supervision on Banks' Technical Efficiency: An Application of
Two-Stage Data Envelopment Analysis", Review of Quantitative
Finance and Accounting, Vol. 30, No. 2.
Pasiouras, F., S. Tanna and C. Zopounidis (2009), "The Impact of Banking
Regulations on Banks' Cost and Profit Efficiency: Cross-Country
Evidence", International Review of Financial Analysis, Vol. 18, No. 5.
Podpiera, R. (2006), "Does Compliance with Basel Core Principles Bring
Any Measurable Benefits?", IMF Staff Papers, Vol. 53, No. 2,
International Monetary Fund, Washington, D.C.
Quah, D. (1996), “Twin peaks: growth and convergence in models of
distribution dynamics”, Economic Journal, Vol. 106.
140

Rioja, F. and N. Valev (2004), "Does One Size Fit All?: A Reexamination of
the Finance and Growth Relationship", Journal of Development
Economics, Vol. 74, No. 2.
Rocha, R., S. Farazi, R. Khouri and D. Pearce (2010a), "The Status of Bank
Lending to Small and Medium Enterprises", MENA Financial Sector
Flagship Report, Improving Access to Finance while Maintaining
Stability in the Middle East and North Africa, World Bank,
Washington, D.C.
Rocha, R., S. Farazi, R. Khouri and D. Pearce (2010b), "The Status of Bank
Lending to SMEs in the Middle and North Africa Region: The
Results of a Joint Survey of the Union of Arab Banks and the World
Bank", World Bank, Washington, D.C.
Rose-Ackerman, S. (1975), "The Economics of Corruption", Journal of Public
Economics, Vol. 4, No. 2.
Saadani, Y., Z. Arvai and R. Rocha (2010), "A Review of Credit Guarantee
Schemes", MENA Financial Sector Flagship Report, Improving
Access to Finance while Maintaining Stability in the Middle East and
North Africa, World Bank, Washington, D.C.
Sapienza, P. (2004), "The Effects of Government Ownership on Bank
Lending", Journal of Financial Economics, Vol. 72, No. 2.
Schumpeter, J.A. (1934), The Theory of Economic Development, Cambridge,
MA: Harvard University Press.
Sealey, C. and J.T. Lindley (1977), “Inputs, Outputs and a theory of
production and cost at depositary financial institutions”, Journal of
Finance, Vol. 32.
Shleifer, A. and R.W. Vishny (1993), "Corruption", Quarterly Journal of
Economics, Vol. 108, No. 3.
Shleifer, A. and R.W. Vishny (1994), "Politicians and Firms", Quarterly
Journal of Economics, Vol. 109, No. 4.
Stiglitz, J.E. (1994), "The role of the state in financial markets", in
Proceedings of the World Bank Annual Conference on Development
Economics, World Bank, Washington, D.C.
Stock, J.H. and M. Yogo (2001), "Testing for Weak Instruments in Linear IV
Regression", unpublished manuscript, Harvard University.
Sundararajan, V., D. Marston and R. Basu (2001), "Financial System
Standards and Financial Stability - The Case of Basel Core
Principles", IMF Working Paper No. 01/62, International Monetary
Fund, Washington, D.C.
141

142 | ANNEX
Tahari, A., P. Brenner, E.D. Vrijer, M. Moretti, A. Senhadji, G.
Sensenbrenner and J. Solé (2007), "Financial Sector Reforms and
Prospects for Financial Integration in Maghreb Countries", IMF
Working Paper No. 07/125, International Monetary Fund,
Washington, D.C.
Tortosa-Ausina, E. (2002), “Exploring efficiency differences over time in
the Spanish banking industry”, European Journal of Operational
Research, Vol. 139, No. 3.
Turk-Ariss, R. (2009), "Competitive Behavior in Middle East and North
Africa Banking Systems", Quarterly Review of Economics and Finance,
Vol. 49, No. 2.
Vander Vennet, R. (2002), "Cost and Profit Efficiency of Financial
Conglomerates and Universal Banks in Europe", Journal of Money,
Credit, and Banking, Vol. 34, No. 1.
Vives, X. (1990), "Trade Association Disclosure Rules, Incentives to Share
Information, and Welfare", RAND Journal of Economics, Autumn, Vol.
21, No. 3.
Wachtel, P. (2001), "Growth and Finance: What Do We Know and How Do
We Know It?", International Finance, Winter, Vol. 4, No. 3.
Wachtel, P. (2003), "How Much Do We Really Know about Growth and
Finance?", Federal Reserve Bank of Atlanta Economic Review, 1 st Quarter,
Vol. 88, No. 1.
Weill, L. (2009), “Convergence in banking efficiency across European
countries”, Journal of International Financial Markets, Institutions &
Money, Vol. 5.
World Bank (2004), "Algeria: Financial Sector Assessment Staff Report",
Financial Sector Assessment Program (FSAP), Washington, D.C.
World Bank (2005), Financial Sector Assessment: A Handbook, jointly
published by the World Bank and the International Monetary Fund,
Washington, D.C.
World Bank (2008), "Egypt: Financial Sector Assessment", Washington,
D.C.
Zhao, T., B. Casu and A. Ferrari (2010), “The impact of regulatory reforms
on cost structure, ownership and competition in Indian banking”,
Journal of Banking and Finance, Vol. 34.
142

ANNEX 1. CEPS SURVEY ON BANKING
SUPERVISION FOR THE SOUTHERN
MEDITERRANEAN COUNTRIES
The survey comprised two distinct parts: data requests and face-to-face
interviews. Data requests (Section I) were sent prior to the interviews to
allow for researchers to develop more focus in the interviews. Face-toface
interviews (Section II) were held with experts in the supervisory
and regulatory agencies, backed with data obtained in section I and
other sources, including a variety of international databases and data
sources from the World Bank, IMF and national sources. The text of the
survey is reproduced below.
SECTION 1. DATA REQUESTS
The following survey is designed to allow the CEPS research team to
better assess the practice of banking supervision in your country. The
data requests cover a variety of regulatory issues, including general
aspects of the supervisory agency (Part 1); licensing, disclosure, and
prudential requirements (Parts 2 to 4); crisis management practices and
schemes (Parts 5 and 6); and, market infrastructure (Part 7). The last
section (Part 8) concludes the interview with final remarks.
For all questions, please respond with the most recent
information. In some cases, space is provided for OPTIONAL data on
prior years. Please enter the data in the corresponding columns to
distinguish between different years.
1. SUPERVISORY AGENCY
1.1. General
1.1.a. Name of agency:
1.1.b. Operational since:
1.1.c. Legal basis (please provide formal legal reference):
1.2. Resources
1.2.a. Budget:
1.2.b. Number of staff (Full-time equivalent (FTE), 2009):
1.2.c. Number of supervisory staff (FTE, 2009):
1.3. Management
1.3.a. Name of the current head of agency:
1.3.b. Current head appointed in:
1.3.c Current head’s term to last until:
| 143

144 | REFERENCES
2. ENTRY AND LICENSING
2.1 General
2.1.a. Name of licensing body:
2.1.b. Operational since:
2.1.c. Legal basis:
2.1.d. New licenses granted:
2.1.e. Number of banks for the following categories:
2.1.e.i. Commercial banks:
2.1.e.ii. Public banks:
2.1.e.iii. Development banks:
2.1.e.iv. Islamic banks:
2.1.e.v. Specialised credit institutions:
2.1.e.vi. Foreign subsidiaries:
2.1.e.vii. Foreign branches:
2.1.f. Total assets of banks for the following categories:
2.1.f.i. Commercial banks:
2.1.f.ii. Public banks:
2.1.f.iii. Development banks:
2.1.f.iv. Islamic banks:
2.1.f.v. Specialised credit institutions:
2.1.f.vi. Foreign subsidiaries:
2.1.f.vii. Foreign branches:
2.1.f.viii. Currency:
2.1.g. Total customer loans of banks for the following categories:
2.1.g.i. Commercial banks:
2.1.g.ii. Public banks:
2.1.g.iii. Development banks:
2.1.g.iv. Islamic banks:
2.1.g.v. Specialised credit institutions:
2.1.g.vi. Foreign subsidiaries:
2.1.g.vii. Foreign branches:
2.1.h. Total customer deposits for the following categories:
2.1.h.i. Commercial banks:
2.1.h.ii. Public banks:
2.1.h.iii. Development banks:
2.1.h.iv. Islamic banks:
2.1.h.v. Specialised credit institutions:
2.1.h.vi. Foreign subsidiaries:
2.1.h.vii. Foreign branches:
2.1.i Concentration of banking sector (assets of top-3 banks as
percentage of total banking assets):
2.2. Entry requirements
2.2.a. Please identify the minimum capital requirements for the
following institutions, whenever applicable:
144

09):
2.2.a.i. Commercial banks:
2.2.a.ii. Public banks:
2.2.a.iii. Development banks:
2.2.a.iv. Islamic banks:
2.2.a.v. Specialised credit institutions:
2.2.a.vi. Foreign subsidiaries:
2.2.a.vii. Foreign branches:
2.2.b. What types of funds may be used as paid-up capital upon entry:
2.2.b.i. Cash?
2.2.b.ii. Government securities?
2.2.b.iii. Borrowed funds?
2.2.b.iv. Other (please specify)?
2.3. Acquisitions
2.3.a. Name of approving body:
2.3.b. Acquisitions granted (2005-2009):
2.4. Rejections
2.4.a. Number of rejected licensing applications for the following
institutions (2005-09):
2.4.a.i. Commercial banks:
2.4.a.ii. Public banks:
2.4.a.iii. Development banks:
2.4.a.iv. Islamic banks:
2.4.a.v. Specialised credit institutions:
2.4.a.vi. Foreign subsidiaries:
2.4.a.vii. Foreign branches:
2.4.b. Number of rejected acquisitions for the following institutions
(2005-09):
2.4.b.i. Commercial banks:
2.4.b.ii. Public banks:
2.4.b.iii. Development banks:
2.4.b.iv. Islamic banks:
2.4.b.v Specialised credit institutions:
2.5. Foreign entities
2.5.a. Licences granted to foreign (majority-owned) branches (2005-09):
2.5.b. Licences granted to foreign (majority-owned) subsidiaries (2005-
2.5.c. Foreign acquisitions granted (2005-09):
3. INFORMATION DISCLOSURE
3.1. General
3.1.a. Number of legal actions taken against auditors for not fulfilling
their responsibilities (2005-09):
3.1.b. Number of legal actions taken against bank directors on failing to
disclose information accurately or truthfully to the public (2005-
09):
145

146 | REFERENCES
4. PRUDENTIAL REQUIREMENTS
4.1. Capital requirements and conditions
4.1.a. What are the minimum capital requirements?
4.1.a.i. Tier 1 ratio (%):
4.1.a.ii. Total (tier 1 + tier 2) capital ratio (%):
4.1.a.iii. Leverage ratio (please specify):
4.1.a.iv. Other (please specify):
4.1.b. What are the actual capital conditions?
4.1.b.i. Tier 1 ratio (%):
4.1.b.ii. Total capital ratio (%):
4.1.b.iii. Simple leverage ratio (%):
Please specify calculation method:
4.1.b.iv. Other:
4.2. Liquidity and diversification requirements
4.2.a. Percentage bank assets held in government bonds (%):
4.2.b. Percentage foreign-currency denominated bank assets (%):
4.2.c. Percentage foreign-currency denominated bank liabilities (%):
4.3. Provisioning rules
4.3.a. Ratio of non-performing loans (NPL) to total loans (%):
4.3.b. How many days is a loan in arrears classified as?
4.3.b.i. Substandard:
4.3.b.ii. Doubtful:
4.3.b.iii. Loss:
4.3.c. What are the minimum provisioning requirements for loans that
are classified as?
4.3.c.i. Substandard:
4.3.c.ii. Doubtful:
4.3.c.iii. Lost:
5. CRISIS MANAGEMENT
5.1. General
5.1.a. Number of times the “cease and desist” actions invoked (2005-09):
5.1.b. Number of banks restructured (2005-09):
5.1.c. Number of banks liquidated (2005-09):
6. DEPOSIT INSURANCE SCHEME
6.1. General
6.1.a. Is there an explicit deposit insurance protection system? If not,
please skip this section.
6.1.b. Name of scheme:
6.1.c. Legal basis:
6.1.d. Operational since:
6.1.e. Number of times depository insurance was activated (2005-09):
6.1.f. How are depositors informed about the scheme?
6.1.f.i. Verbally when opening accounts:
6.1.f.ii. Notices or brochures available at branches:
146

6.1.f.iii. Media announcements:
6.1.f.iv. Direct mailings:
6.1.f.v. Publication in official journal/newspaper:
6.1.f.vi. Other (please specify):
6.2. Protection
6.2.a. What is the insurance limit per account for:
6.2.a.i. Local currency accounts?
6.2.a.ii. Foreign currency accounts, if covered?
6.2.b. What is the insurance limit per person for:
6.2.b.i. Local currency deposits?
6.2.b.ii. Foreign currency accounts, if covered?
6.2.c. Is there a ‘co-insurance mechanism’? In other words, are
depositors insured for some part of their deposits,
notwithstanding the insurance limits above? If so, please specify
the applicable ratio. Otherwise, leave blank.
6.2.d. Is there a legal period for which all the covered depositors must
be paid in full? If so, specify the maximum allowed delay:
6.3. Funding
6.3.a. Please check how often are premiums collected:
6.3.a.i. Other (please specify):
6.3.b. If there is an ex-ante fund, please specify the amount of funds
accumulated as percent of total customer deposits (2005-09):
7. MARKET INFRASTRUCTURE
7.1. Liquidity management
7.1.a. Agency responsible for implementing monetary policy:
7.1.b. Legal basis:
7.1.c. Operational since:
7.1.d. Amount of transactions in the inter-bank market:
7.1.e. Please detail which monetary policy tools are available for use:
7.1.e.i. Interest rate ceilings or credit ceilings? If so, please specify
requirements:
7.1.e.ii. Reserve requirements i.e. deposit reserve requirement
ratio? If so, please specify requirements:
7.1.e.iii. Liquid asset restrictions/ratios? If so, please specify
requirements:
7.1.e.iv. Standing facilities? If so, amount of lending (2009):
7.1.e.v. Open market operations, i.e. buying/selling assets on
secondary market, repo, or foreign exchange markets? If so,
amount of lending (2009):
7.1.e.vi. Government or central bank security auctions? If so, please
specify methods (volume tenders; interest rate bids;
multiple-rate auctions; etc.). If so, amount of lending (2009):
7.1.e.vii. Others? (please specify as above)
147

148 | REFERENCES
7.1.f. Which of the above tools has been the most predominant one over
the past five years? Please support the argument with evidence,
i.e. amount/volume of transactions, etc.
7.2. Payment system
7.2.a. Number and value of card payments:
7.2.b. Number and value of cheques drawn:
7.2.c. Number and value of ATM withdrawals:
7.3. Public debt management
7.3.a. What percent of domestic credit is available to the
7.3.a.i. general government (2009)?
7.3.a.ii. state-owned (public) enterprises (2009)
SECTION II. FACE-TO-FACE INTERVIEW
The following interview is designed to allow the CEPS research team to
better assess the practice of banking supervision in your country. The
questions address a variety of regulatory issues, including general
aspects of the supervisory agency (Part 1); licensing, disclosure, and
prudential requirements (Parts 2 to 4); crisis management practices and
schemes (Parts 5 and 6); and, market infrastructure (Part 7). The last
section (Part 8) concludes the interview with final remarks.
Please answer the questions fully, providing details and
references, whenever applicable. You will have an opportunity to make
additional comments at the end of the questionnaire.
1. SUPERVISORY AGENCY
1.1. General
1.1.a. What are the official roles and aims of the supervisory agency?
1.1.b. How is the agency financed? By public funds, premiums on financial
institutions, or else?
1.2. Resources
1.2.a. What are the educational requirements for supervisors, including
required diplomas, specialisations, etc.?
1.2.b. Approximately what percent of the agency’s budget is devoted to
training programmes for supervisors?
1.2.c. Please comment on existing or potential cooperation opportunities with
the EU to improve the resource utilisation of the agency, including
technical aid, training, and twinning programmes.
1.2.d. Has the supervisory agency signed any mutual recognition agreements
or memorandum of understanding (MoU) with the EU member states?
If so, what are the details of these agreements?
148

1.2.e. Do the supervisors meet regularly with EU’s supervisory authorities?
Please provide details on frequency and types of meetings.
1.2.f. Does the supervisor agency have a seat as
observer in Committee of European Banking
Supervisors (CEBS)?
1.3. Management
1.3.a. Who appoints the head of agency?
YES/NO
1.3.b.i. If there is a fixed term of service for the head of agency, what is the
term?
1.3.b.ii. Can the fixed-term be renewed? YES/NO
1.3.c. Have there been instances of dismissal of the supervisory agency’s
management over the past few years? If so, please provide details.
1.4. Liability
1.4.a. Can the agency or management be held legally
liable for its actions?
YES/NO
1.4.b. Have the agency or management been held liable for its actions over the
past few years? If so, please provide details on the case and final ruling.
2. ENTRY AND LICENSING
2.1. Entry requirements
2.1.a. What is the maximum delay for fulfilling minimum capital
requirements? Are delays tolerated in fulfilling these requirements?
2.2. Rejections
2.2.a. Are the grounds for rejecting a licensing application specified by law? If
so, what are these criteria?
2.2.b. If applicable, what are the most typical reasons for rejecting licensing
applications?
2.2.c. If applicable, what are the most typical reasons for rejecting
acquisitions?
2.3 Foreign entities
2.3.a. Is there a political commitment to manage foreign entry into banking
sector? Please comment on the arguments, potential benefits and harms
that unmanaged foreign entry may pose.
2.3.b. How are the home-host supervisory conflicts resolved for foreign
financial institutions established in your country? Is there an active
cooperation with EU’s supervisors?
3. INFORMATION DISCLOSURE
3.1 External audit
3.1.a. Please name the top-three certified auditors in terms of number of banks
149

150 | REFERENCES
150
audited:
3.1.b. Please comment on the adequacy and quality of external audits. In
particular, do the auditors face difficulties (lack of training, resources,
experience, etc.) in making an independent and fair judgement? If so,
please identify how these issues can be overcome in the future?
3.2 Financial statements
3.2.a. Are there delays (i.e. more than 1-2 days) in sending audited financial
statements to authorities? If so, what are the typical reasons for filing
late?
3.2.b. Are banks’ directors legally liable for the
accuracy and truthfulness of information
disclosed to the public or to the authorities?
4. PRUDENTIAL REQUIREMENTS
4.1. Capital requirements
4.1.a. Are the minimum requirements anticipated to remain
the same over the next few years?
YES/NO
YES/NO
4.1.b. Does credit to SME’s receive any preferential treatment? YES/NO
4.2. Liquidity and diversification requirements
4.2.a. What are the rules and guidelines on asset diversification (maximum
exposure to a single borrower, minimum diversification of loans among
sectors, concentration limits, etc.)?
4.2.b. What is the reserve requirement, i.e. the minimum deposit reserve
requirement ratio?
4.3 Basel implementation
4.3.a. Is the Basel II framework currently applicable? If so, since when? If not,
when is it anticipated to be applicable?
4.3.b Are there any differences in the implementation of Basel II for public
banks and private banks? If so, what are the official reasons for making
such a distinction?
4.3.c. What is the definition of tier I and tier II capital?
4.3.c. Are there plans to implement Basel III rules? If so, by when?
4.3.d. Is the internal ratings based (IRB) approach applicable? If so, since
when? If not, when is it anticipated to be applicable?
4.3.e. Please comment on the main difficulties in applying Basel II and the
upcoming Basel III.
4.3.f. Please comment on whether Basel II puts domestic banks at a
disadvantage in comparison to foreign-owned banks.
4.3.g. Please comment on existing or potential cooperation opportunities with

the EU to improve Basel II implementation, including technical aid,
training, and twinning programmes.
4.4. Provisioning rules
4.4.a. When is a loan classified as a “non-performing loan” (NPL)?
4.4.b. Is a high NPL ratio a challenge for the banking sector? If so, what are the
relevant policy responses to combat with this challenge?
5. CRISIS MANAGEMENT
5.1. General
5.1.a. Is there a regime exclusively applicable for bank insolvencies? If so,
please detail the legal basis for the regime.
5.1.b. Please comment to what extent the current laws and structures ensure a
speedy resolution/restoration of troubled banks?
5.1.c. What is the level of cooperation between the supervisory authority and
government in resolution/restoration of troubled banks?
5.1.d. Has the financial crisis had any impact on banks in your country? If so,
what was the impact? Increased NPL, liquidity support, insolvencies,
etc.? Please provide details.
5.2. Prompt corrective action
5.2.a. Which body or bodies have the power to
5.2.a.i. Order a bank to “cease and desist” all activities? (please specify below)
5.2.a.ii. Impose penalties due to infraction of “cease and desist” orders?
5.2.a.iii. Override management’s or director’s decisions?
5.3. Restructuring
5.3.a. Which of the following actions are available during restructuring:
5.3.a.i. Mergers YES/NO
5.3.a.ii. Recapitalisation YES/NO
5.3.a.iii. Outright sale of the bank YES/NO
5.3.a.iv. Acquisition of assets by another bank YES/NO
5.3.a.v. Transfer of assets to an asset management company YES/NO
5.3.a.vi. Mergers YES/NO
5.3.b. Which body or bodies have the power to:
5.3.b.i. Supersede shareholders’ rights?
5.3.b.ii. Replace management and directors?
5.3.b.iii. Forebear (i.e. withhold the application of) certain regulations?
5.3.c. To what extent are shareholders’ rights upheld during restructuring,
especially for those that contribute to the recapitalisation of the bank?
151

152 | REFERENCES
5.3.d. Is there a difference in the treatment of public banks
and commercial banks during restructuring?
5.4. Liquidation
5.4.a. Which body or bodies have the power to:
5.4.a.i. Declare a bank insolvent? (please identify below)
5.4.a.ii. Appoint a receiver or liquidator?
5.4.a.iii. Terminate shareholders’ rights?
6. DEPOSIT INSURANCE SCHEME
6.1. General
6.1.a. Is there an explicit deposit insurance protection
system? If not, please skip to section 7.
6.1.b. Does the system provide coverage for foreignowned
banks?
6.2. Protection
152
YES/NO
YES/NO
YES/NO
6.2.a. Is there a legally determined maximum delay to pay covered depositors
in full in the event of the failure of a bank? If so, what is the period?
6.2.b. Please comment on the anticipated delay to pay the covered depositors
in full in the event of the failure of an average-sized bank. Is this in line
with the legal period mentioned above?
6.2.c. Historically, have depositors been compensated fully up to the legal
protection? If not, please comment on the reasons, i.e. shortage of exante
funds, foreign exchange volatility, etc., for why the actual
protection fell short of the intended protection.
6.2.d. Please comment on the powers of the insurer in intervening in a bank,
revoking deposit insurance, taking legal action against management, etc.
6.3. Funding
6.3.a. Are premiums risk-adjusted? If so, please comment on how the risk
ratings or valuations are determined.
6.3.b. Is there an ex-ante fund or an ex-post system? If the latter, please detail
how the system is to operate in the event of a trouble? Is the system
backed by government?
7. MARKET INFRASTRUCTURE
7.1. Liquidity management
7.1.a. How does the monetary policy impact banking in your country
currently? What are the relevant anticipated changes in monetary policy
in upcoming years? How will these developments impact banking?
7.1.b. Please comment on existing or potential cooperation arrangements with
the EU to improve the use of monetary policy tools, including technical

aid, training, and twinning programmes.
7.2. Foreign exchange policy
7.2.a. How relevant are foreign currency risks in banking? How are these risks
likely to evolve with the introduction of new exchange policies in
upcoming years, if applicable?
7.3. Payment system
7.3.a. What is the predominance of cash-based transactions as opposed to card
payments or paperless transactions?
7.3.b. Is there a real-time gross settlement system? If so, since when has it been
operational?
7.4 Institutional environment
7.4.a. Please comment on the effectiveness of the judicial system on loan
recovery. What needs to be done to overcome challenges, if any?
7.4.b. How effective are laws in protecting private property rights? What
needs to be done to overcome challenges, if any?
7.4.c. Please comment on the degree of government involvement in banking.
Is there an active government policy to direct credit? rovide details.
8. FINAL REMARKS
8.1. Please comment on the key challenges for banking supervision in your
country, distinguishing between challenges faced by policy-makers,
banking sector, auditors, investors and depositors.
8.2. What are the most important regulatory and supervisory developments
that have taken place in the past few years? What are the anticipated
changes in the upcoming years?
8.3. Please comment on the general cooperation opportunities with the EU to
improve banking supervision.
153